def findall2(all_trees,all_files):
#show 10 candidates
k=10
for files in all_files:
for f in files:
temp={'ds':[],'ids':[],'files':[]}
fsg=getHash3(f)
for tree in all_trees:
ds,ids=tree.query(fsg,k)
temp['ds'].append(ds)
temp['ids'].append(ids)
temp['files'].append(files)
id_sort=np.argsort(temp['ds'])
sort_d=temp['ds'][id_sort[0:k]]
sort_ids=temp['ids'][id_sort[0:k]]
sort_files=temp['files'][id_sort[0:k]]
plt.figure(1)
for i in xrange(k):
print "%03d dis: %.3f %s"%(sort_ids[i],sort_d[i],sort_files[i])
plt.subplot(2,5,i+1)
plt.title("%03d dis: %.3f"%(sort_ids[i],sort_d[i]),fontsize=10)
im_icon=np.array(Image.open(sort_files[i]).convert('L'))
plt.imshow(im_icon)
plt.axis('off')
plt.set_cmap('gray')
plt.show()
plt.ginput(1)
def main(base_dir):
BASE_DIR = base_dir
# Load the set of pictures
ic = io.ImageCollection(BASE_DIR + '*.JPG')
# Select points on the first picture
f, ax = plt.subplots(1,1)
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
ax.autoscale(enable=True, axis='both', tight=True);
plt.tight_layout(pad=0.4, w_pad=0.0, h_pad=0.0)
ax.imshow(ic[0])
coords = [plt.ginput(8, timeout=0)]
plt.close()
# Load first picture side-by side with second, select points.
# Scroll through images one-by-one
for i, img in enumerate(ic[1:]):
ax1 = plt.subplot2grid((6,10),(0,1), rowspan=6, colspan=9)
ax0 = plt.subplot2grid((6,10),(0,0))
for ax in [ax0, ax1]:
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.tight_layout(pad=0.4, w_pad=0.0, h_pad=0.0)
#f, (ax0,ax1) = plt.subplots(1,2)
ax0.imshow(ic[i])
for coord in coords[i]:
ax0.scatter(coord[0],coord[1])
ax1.imshow(img)
coords.append(plt.ginput(8, timeout=0))
plt.close()
# Use a similarity transformation to transform each one.
if not os.path.exists(BASE_DIR + 'corrected'):
os.mkdir(BASE_DIR + 'corrected')
np.save(BASE_DIR + 'corrected/coords.npy', coords)
io.imsave(BASE_DIR + 'corrected/0.jpg', ic[0])
for i, img in enumerate(ic[1:]):
tf = transform.estimate_transform('similarity', np.array(coords[0]), np.array(coords[i+1]))
# Use a translation transformation to center both images for display purposes
img_warped = transform.warp(img, inverse_map=tf,
output_shape=(1728,3072))
print BASE_DIR + 'corrected/%d.jpg' %(i+1)
print img_warped
io.imsave(BASE_DIR + 'corrected/%d.jpg' %(i+1), img_warped)
def selectStartGoalPoints(ax, img):
print 'Select a starting point'
ax.set_xlabel('Select a starting point')
occupied = True
while(occupied):
point = ppl.ginput(1, timeout=-1, show_clicks=False, mouse_pop=2)
start = [ round(point[0][0]), round(point[0][1]) ]
if(img[int(start[1])][int(start[0])][0] == 255):
occupied = False
ax.plot(start[0], start[1], '.r')
else:
print 'Cannot place a starting point there'
ax.set_xlabel('Cannot place a starting point there, choose another point')
print 'Select a goal point'
ax.set_xlabel('Select a goal point')
occupied = True
while(occupied):
point = ppl.ginput(1, timeout=-1, show_clicks=False, mouse_pop=2)
goal = [ round(point[0][0]), round(point[0][1]) ]
if(img[int(goal[1])][int(goal[0])][0] == 255):
occupied = False
ax.plot(goal[0], goal[1], '.b')
else:
print 'Cannot place a goal point there'
ax.set_xlabel('Cannot place a goal point there, choose another point')
ppl.draw()
return start, goal
def show(self):
import matplotlib.pyplot as plt
print("number of images: {}".format(len(self.I)))
for i in xrange(len(self.jsonfiles)):
f=open(self.jsonfiles[i],"r")
js=json.loads(f.read())
f.close()
##init and show
img_path=''
if js['path'][0:2]=='./':
img_path= rootdir + js['path'][1:]
elif js['path'][0]=='/':
img_path= rootdir + js['path']
else:
img_path= rootdir + '/' +js['path']
print(img_path)
im=np.array(Image.open(img_path).convert('L'))
plt.hold(False)
plt.imshow(im)
plt.hold(True)
for j in range(self.size):
#samples[x]=[0_class,1_img, 2_row, 3_column]^T
if self.samples[1,j]==i:
plt.text(self.samples[3,j], self.samples[2,j], "%03d"%self.samples[0,j], fontsize=12,color='red')
#plt.plot(self.samples[3,j],self.samples[2,j],markers[self.samples[0,j]])
plt.set_cmap('gray')
plt.show()
plt.ginput(1)
plt.close('all')
def manual_sync_pick(flow, gyro_ts, gyro_data):
# First pick good points in flow
plt.clf()
plt.plot(flow)
plt.title('Select two points')
selected_frames = [int(round(x[0])) for x in plt.ginput(2)]
# Now pick good points in gyro
plt.clf()
plt.subplot(211)
plt.plot(flow)
plt.plot(selected_frames, flow[selected_frames], 'ro')
plt.subplot(212)
plt.plot(gyro_ts, gyro_data.T)
plt.title('Select corresponding sequence in gyro data')
plt.draw()
selected = plt.ginput(2) #[int(round(x[0])) for x in plt.ginput(2)]
gyro_idxs = [(gyro_ts >= x[0]).nonzero()[0][0] for x in selected]
plt.plot(gyro_ts[gyro_idxs], gyro_data[:, gyro_idxs].T, 'ro')
plt.title('Ok, click to continue to next')
plt.draw()
plt.waitforbuttonpress(timeout=10.0)
plt.close()
return (tuple(selected_frames), gyro_idxs)
def trainZone(self,event): if str(event.keysym) == 'space': flag, frame = self.cap.read() if flag == True: #~ grayFrame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) plt.imshow(frame) #~ self.mask = np.zeros(frame.shape[:2],dtype = 'uint8') pnt1 = np.int32(plt.ginput(n=0, timeout=-1, show_clicks=True)) self.pnt1 = pnt1.reshape((-1,1,2)) xMax = max(self.pnt1[:,:,0])[0] yMax = max(self.pnt1[:,:,1])[0] xMin = min(self.pnt1[:,:,0])[0] yMin = min(self.pnt1[:,:,1])[0] cv2.polylines(frame,[self.pnt1],True,(0,255,255),thickness=3) plt.imshow(frame) pnt2 = np.int32(plt.ginput(n=0, timeout=-1, show_clicks=True)) self.pnt2 = pnt2.reshape((-1,1,2)) xMax = max(self.pnt2[:,:,0])[0] yMax = max(self.pnt2[:,:,1])[0] xMin = min(self.pnt2[:,:,0])[0] yMin = min(self.pnt2[:,:,1])[0] cv2.polylines(frame,[self.pnt2],True,(0,255,255),thickness=3) plt.imshow(frame)
def solver(N):
for i in range(N):
for j in range(N):
vic, tab = maneja_reinas(reinas, 'Agregar reina', i, j)
img_reinas.set_data(tab)
plt.draw()
plt.ginput()
if vic:
fig.text(
0.5, 0.5, 'Felicitaciones!', color='orange', size='xx-large',
family='humor sans', ha='center', va='center', alpha=0.5,
bbox=dict(facecolor='ivory', edgecolor='orange', boxstyle='round')
)
plt.draw()
plt.ginput()
if tab[i][j]==4:
if i in range(N) and j in range(N):
vic, tab = maneja_reinas(reinas, 'Agregar reina', i, j)
j=j+1
print ("{}{}".format(i,j))
def get_point(im1, im2):
plt.imshow(im1)
im1_pt = plt.ginput()[0]
plt.close()
plt.imshow(im2)
im2_pt = plt.ginput()[0]
plt.close()
return im1_pt,im2_pt
def get_points(im1, im2):
print('Please select 2 points in each image for alignment.')
plt.imshow(im1)
p1, p2 = plt.ginput(2)
plt.close()
plt.imshow(im2)
p3, p4 = plt.ginput(2)
plt.close()
return (p1, p2, p3, p4)
def tps_rpm(x_nd, y_md, n_iter = 5, reg_init = .1, reg_final = .001, rad_init = .2, rad_final = .001, plotting = False, verbose=True, f_init = None):
n,d = x_nd.shape
regs = loglinspace(reg_init, reg_final, n_iter)
rads = loglinspace(rad_init, rad_final, n_iter)
f = ThinPlateSpline.identity(d)
for i in xrange(n_iter):
if f.d==2 and i%plotting==0:
import matplotlib.pyplot as plt
plt.clf()
if i==0 and f_init is not None:
xwarped_nd = f_init(x_nd)
print xwarped_nd.max(axis=0)
else:
xwarped_nd = f.transform_points(x_nd)
# targ_nd = find_targets(x_nd, y_md, corr_opts = dict(r = rads[i], p = .1))
corr_nm = calc_correspondence_matrix(xwarped_nd, y_md, r=rads[i], p=.2)
wt_n = corr_nm.sum(axis=1)
targ_nd = np.dot(corr_nm/wt_n[:,None], y_md)
# if plotting:
# plot_correspondence(x_nd, targ_nd)
f.fit(x_nd, targ_nd, regs[i], wt_n = wt_n, angular_spring = regs[i]*200, verbose=verbose)
if plotting and i%plotting==0:
if f.d==2:
plt.plot(x_nd[:,1], x_nd[:,0],'r.')
plt.plot(y_md[:,1], y_md[:,0], 'b.')
pred = f.transform_points(x_nd)
if f.d==2:
plt.plot(pred[:,1], pred[:,0], 'g.')
if f.d == 2:
plot_warped_grid_2d(f.transform_points, x_nd.min(axis=0), x_nd.max(axis=0))
plt.ginput()
elif f.d == 3:
from lfd import warping
from brett2.ros_utils import Marker
from utils import conversions
Globals.setup()
mins = x_nd.min(axis=0)
maxes = x_nd.max(axis=0)
mins[2] -= .1
maxes[2] += .1
Globals.handles = warping.draw_grid(Globals.rviz, f.transform_points, mins, maxes, 'base_footprint', xres=.1, yres=.1)
orig_pose_array = conversions.array_to_pose_array(x_nd, "base_footprint")
target_pose_array = conversions.array_to_pose_array(y_md, "base_footprint")
warped_pose_array = conversions.array_to_pose_array(f.transform_points(x_nd), 'base_footprint')
Globals.handles.append(Globals.rviz.draw_curve(orig_pose_array,rgba=(1,0,0,1),type=Marker.CUBE_LIST))
Globals.handles.append(Globals.rviz.draw_curve(target_pose_array,rgba=(0,0,1,1),type=Marker.CUBE_LIST))
Globals.handles.append(Globals.rviz.draw_curve(warped_pose_array,rgba=(0,1,0,1),type=Marker.CUBE_LIST))
f.corr = corr_nm
return f
def checkSignals(self):
from matplotlib.pyplot import plot, figure, close, ginput
for i in range(0, len(self.Signals)):
print('Signal ' + str(i))
s = self.Signals[i]
if not self.isGoodSignal(s, self.SamplingPeriod):
figure()
plot(s)
ginput(timeout=0)
close("all")
def plot_cost(grid):
import matplotlib.pyplot as plt
import scipy.stats as ss
plt.clf()
#plot_arr = ss.rankdata(grid.array).reshape(grid.array.shape)
plot_arr = grid.array
print grid.array.max(), grid.array.min()
plt.imshow(plot_arr, extent = [grid.xticks[0], grid.xticks[-1], grid.yticks[0], grid.yticks[-1]])
plt.title("cost")
print "click on the plot to continue"
plt.ginput()
def _ginput(*args, **kwargs):
"""
Hides the stupid default warnings when using ginput. There has to be
a better way...
"""
if MatplotlibDeprecationWarning is None:
return plt.ginput(*args, **kwargs)
else:
warnings.simplefilter('ignore', MatplotlibDeprecationWarning)
out = plt.ginput(*args, **kwargs)
warnings.warn('default', MatplotlibDeprecationWarning)
return out
def pickloc(ax=None, zoom=10):
"""
:INPUTS:
ax : (axes instance) -- axes in which to pick a location
zoom : int -- zoom radius for target confirmation
: 2-tuple -- (x,y) radii for zoom confirmation.
"""
# 2011-04-29 19:26 IJC:
# 2011-09-03 20:59 IJMC: Zoom can now be a tuple; x,y not cast as int.
pickedloc = False
if ax is None:
ax = plt.gca()
axlimits = ax.axis()
if hasattr(zoom, '__iter__') and len(zoom)>1:
xzoom, yzoom = zoom
else:
xzoom = zoom
yzoom = zoom
while not pickedloc:
ax.set_title("click to select location")
ax.axis(axlimits)
x = None
while x is None:
selectevent = plt.ginput(n=1,show_clicks=False)
if len(selectevent)>0: # Prevent user from cancelling out.
x,y = selectevent[0]
#x = x.astype(int)
#y = y.astype(int)
if zoom is not None:
ax.axis([x-xzoom,x+xzoom,y-yzoom,y+yzoom])
ax.set_title("you selected xy=(%i,%i)\nclick again to confirm, or press Enter/Return to try again" %(x,y) )
plt.draw()
confirmevent = plt.ginput(n=1,show_clicks=False)
if len(confirmevent)>0:
pickedloc = True
loc = confirmevent[0]
return loc
def get_all_linepoints_user_input(ImgName, DirToSave, NumPointVecs=45):
# user input to select lines for enhanced measurement detection
Img = cv2.imread(ImgName, 0)
if not os.path.isfile(ImgName):
raise Exception("Accumulated image not defined or not on right path", "raised in get_all_linepoints_user_input")
cOrigImg = cv2.cvtColor(Img, cv2.COLOR_GRAY2BGR)
if not os.path.isfile(DirToSave + "/LinePoints.p"):
print "pick lines"
plt.ion()
Lines = []
for i in range(0, NumPointVecs / 3):
while True:
cImg = cOrigImg.copy()
TryAgain = []
plt.title("Please click 3 lines")
plt.imshow(cImg)
RawLineUV = plt.ginput(6)
LineUV = [RawLineUV[i:i + 2] for i in range(0, len(RawLineUV), 2)]
if len(LineUV) == 3:
# convert ot int
for CurLineUV in LineUV:
LineUVShow = [(int(x[0]), int(x[1])) for x in CurLineUV]
cv2.line(cImg, LineUVShow[0], LineUVShow[1], (255, 0, 0), 2)
plt.title("click on lower half of image to accept, on upper to neglect")
plt.imshow(cImg)
# click on image to do anew
UV = plt.ginput(1)
if not not UV and UV[0][1] > Img.shape[0] / 2:
break
Lines.extend(LineUV)
cv2.line(cOrigImg, LineUVShow[0], LineUVShow[1], (255, 0, 0), 2)
pickle.dump(Lines, open(DirToSave + "/LinePoints.p", "wb"))
print "saved lines as " + DirToSave + "/LinePoints.p"
else:
print "!!!!!!!!!!!!!! loading lines !!!!!!!!!!!!!!!!!"
Lines = pickle.load(open(DirToSave + "/LinePoints.p", "rb"))
cImg = cOrigImg.copy()
for Line in Lines:
CurLine = [(int(x[0]), int(x[1])) for x in Line]
cv2.line(cOrigImg, CurLine[0], CurLine[1], (255, 0, 0), 2)
plt.imshow(cOrigImg)
plt.ioff()
print "please close the window"
plt.show()
# sort left and right
return sorted(Lines, key=lambda x: x[0][0]), cImg
def findall(tree,all_files):
for f in all_files:
fsg=getHash3(f)
d,ids=tree.query(fsg,k=10)
plt.figure(1)
for i,index in enumerate(ids):
print "%03d dis: %.3f %s"%(index,d[i],all_files[index])
plt.subplot(2,5,i+1)
plt.title("%03d dis: %.3f"%(index,d[i]),fontsize=10)
im_icon=np.array(Image.open(all_files[index]).convert('L'))
plt.imshow(im_icon)
plt.axis('off')
plt.set_cmap('gray')
plt.show()
plt.ginput(1)
def cut_bounds(self, data):
def tellme(s):
print s
plt.title(s,fontsize=16)
plt.draw()
plt.clf()
plt.plot(data)
plt.setp(plt.gca(),autoscale_on=False)
tellme('You will select start and end bounds. Click to continue')
plt.waitforbuttonpress()
happy = False
while not happy:
pts = []
while len(pts) < 2:
tellme('Select 2 bounding points with mouse')
pts = plt.ginput(2,timeout=-1)
if len(pts) < 2:
tellme('Too few points, starting over')
time.sleep(1) # Wait a second
plt.fill_between(x, y1, y2, alpha=0.5, facecolor='grey', interpolate=True)
tellme('Happy? Key click for yes, mouse click for no')
happy = plt.waitforbuttonpress()
bounds = sorted([int(i[0]) for i in pts])
plt.clf()
print bounds
return data[bounds[0] if bounds[0] > .02*len(data) else 0:bounds[1] if bounds[1] < len(data)-1 else len(data)-1],bounds
def callback(self, img, depth):
## Get image as numpy array
rgb = self.bridge.imgmsg_to_cv(img)
img = np.asarray(rgb)
self.marker_finder.findmarker(img)
detections = self.marker_finder.detections()
if len(detections) != 2:
return
vis = self.marker_finder.draw_detections(img)
depth_16 = np.asarray(
self.bridge.imgmsg_to_cv(depth, desired_encoding="passthrough"))
depth = depth_16 / np.float32(1000.)
lpt, rpt = [c.astype(np.int) for c, r, _ in detections]
if np.max(depth_16) == 0:
return
lz = depth[lpt[1], lpt[0]]
rz = depth[rpt[1], rpt[0]]
if lz == 0 or rz == 0 or self.Kinv is None:
return
plt.clf()
plt.imshow(vis)
_, campt, _ = plt.ginput(3, timeout=-1)
camz = depth[campt[0], campt[1]]
print "Left marker", self.pt3d_from_imgpt(lpt, lz)
print "Right marker", self.pt3d_from_imgpt(rpt, rz)
def GINPUT_ROUTINE(image, selected_pointsrow, selected_pointscolumn):
'''INPUT:
image = where points are supposed to be selected
OUTPUT:
image image with modified values
'''
PLOT_IMAGE(image)
print ("Please selected 2 points")
pts = plt.ginput(n=2, timeout=10)
pts = np.array(pts)
c = int(pts[0][0])
r = int(pts[0][1])
c1 = int(pts[1][0])
r1 = int(pts[1][1])
selected_pointsrow.append(r)
selected_pointsrow.append(r1)
selected_pointscolumn.append(c)
selected_pointscolumn.append(c1)
modified_pixeel_value = image[int(r)][int(c)]
for row in range(r, r1):
for col in range(c, c1):
image[row][col] = modified_pixeel_value
# plt.show()
return image, selected_pointsrow, selected_pointscolumn
def get_windows(image):
"""Display the given image and record user selected points.
Parameters
----------
image : M,N,3 ndarray
The image to be displayed.
Returns
-------
array : n_points,2
An array of coordinates in the image. Each row corresponds to the x, y
coordinates of one point. If an odd number of points are specified, the
last one will be discarded.
"""
plt.interactive(True)
plt.imshow(image)
plt.show()
crop = plt.ginput(0)
plt.close()
plt.interactive(False)
# remove last point if an odd number selected
crop = crop[:-1] if np.mod(len(crop), 2) else crop
return np.vstack(crop).astype('int')[:, [1, 0]]
def play(self):
while True:
pos = plt.ginput(1,timeout=-1)[0]
node = self.which_clicked(pos)
if node is None:
continue
elif node == self.position[0]:
# clicking on self will undo last move
if self.previous_positions != []:
self.update( self.previous_positions.pop() )
self.switch_colors(antisense=True)
self.draw()
else:
if node in self.next_pos:
self.previous_positions.append(self.position)
self.update( self.next_pos[node] )
self.switch_colors()
self.draw()
if self.position == self.states_graph.goal:
break
self.ax.set_title("Congrats, you win !")
self.ax.figure.canvas.draw()
def getCorrespondence(imageA, imageB):
getAuto = int(1);
# num = 8;
num = 20;
if(getAuto == 0):
# Display images, select matching points
fig = plt.figure()
figA = fig.add_subplot(1,2,1)
figB = fig.add_subplot(1,2,2)
# Display the image
# lower use to flip the image
figA.imshow(imageA)#,origin='lower')
figB.imshow(imageB)#,origin='lower')
plt.axis('image')
# n = number of points to read
pts = plt.ginput(n=num, timeout=0)
print(pts);
pts = np.reshape(pts, (2, num/2, 2))
print(pts);
return pts
# mountain
pts = np.array([(290.99673461310101, 168.49247971502086), (351.77211557490591, 132.47743914506236), (381.03433603799715, 139.23025925192962), (455.31535721353646, 161.7396596081536), (471.07193746289317, 148.23401939441919), (511.58885810409652, 175.24529982188801), (507.08697803285168, 359.82238274292507), (403.54373639422113, 319.30546210172179), (390.03809618048672, 341.81486245794582), (290.99673461310101, 359.82238274292507), (5.9867999208391893, 139.23025925192951), (80.267821096378498, 109.96803878883827), (107.27910152384732, 121.22273896695026), (177.05824262814178, 161.73965960815349), (195.06576291312103, 148.23401939441908), (224.32798337621227, 175.2452998218879), (208.57140312685544, 353.06956263605764), (98.275341381357748, 303.54888185236484), (82.518761132000918, 326.05828220858882), (3.7358598852168825, 353.06956263605764)])
# tower 11_22
# pts = np.array([(749.08560430737248, 492.08751712086166), (576.71815519765732, 604.87115666178647), (774.62152269399712, 628.27908184952548), (908.68509422377565, 536.7753742974545), (294.87512870164869, 455.91163273981022), (107.61172719973592, 568.69527228073491), (311.89907429273171, 592.10319746847404), (445.96264582251024, 506.98346951305916)])
# tower 22_11
# pts = np.array([(295.82305294478812, 458.03962593869562), (104.30366504510459, 570.82326547962043), (310.71900533698573, 589.97520426958886), (451.16655646342036, 464.42360553535173), (746.0096868653477, 492.08751712086155), (575.77023095451796, 606.99914986067165), (773.67359845085775, 626.15108865063996), (911.99315637840664, 494.21551031974684)])
pts = np.reshape(pts, (2, num/2, 2))
return pts
def genData_interactive(numClasses,numClicks):
'''
Generate a 2D data set interactively by letting the user click on the axes to distribute the points. Each click generates 20 points randomly sampled from an isotropic Gaussian. Multiple classes are possible. numClicks it the number of clicks per class.
Data set array X is returned (numClasses*numClicks x 2) and Y Classes vector
'''
kPer = 20
kSigma = [[0.3,0.0],[0.0,0.3]]
plt.figure()
plt.axis([-10,10,-10,10])
plt.show()
X = np.zeros([kPer*numClicks*numClasses,2])
Y = np.zeros([kPer*numClicks*numClasses],dtype=np.int16)
for c in range(numClasses):
for i in range(numClicks):
plt.title('Sampling Clicks For Class %d/%d, Click %d/%d'%(c,numClasses,i,numClicks))
x = np.squeeze(plt.ginput(1))
xsamples = genData_gaussian(x,kSigma,kPer)
current = c*kPer*numClicks + i*kPer
X[current:current+kPer,:] = xsamples
Y[current:current+kPer ] = c*np.ones([kPer],dtype=np.int16)
plt.plot(xsamples[:,0],xsamples[:,1],kDraw[c%len(kDraw)],hold=True)
plt.title('Done Sampling Data')
return (X,Y)
def mcmc_explore(self,niter=30,stepsize=0.5):
"""Explore the space of PSFs."""
current_pos = self.lle_proj[0]
current_density = self.lle_density(current_pos)
print("Computing overall background density for plotting...")
density, extent = self.display_lle_space(return_density=True)
for i in range(niter):
plt.clf()
plt.subplot(121)
jump = np.random.normal(size=self.ndim)*stepsize*self.h_density
trial = current_pos + jump
new_density = self.lle_density(trial)
if new_density/current_density > np.random.random():
current_density = new_density
current_pos = trial
psf = self.find_lle_psf(current_pos, return_image=True)
plt.imshow(np.arcsinh(psf/np.max(psf)/0.01),interpolation='nearest', cmap=cm.cubehelix)
plt.title("lnprob: {0:5.1f}".format(current_density))
plt.subplot(122)
plt.imshow(density,extent=extent,cmap=cm.gray)
plt.plot(self.tri.points[:,0],self.tri.points[:,1],'b.')
plt.title("Pos: {0:5.2f} {1:5.2f}".format(current_pos[0], current_pos[1]))
plt.plot(current_pos[0], current_pos[1], 'ro')
plt.axis(extent)
plt.draw()
#!!! Problem: Current matplotlib does not draw here. !!!
dummy = plt.ginput(1)
return None
def decasteljau_interactive(n):
""" Makes an interactive plot using a slider bar to show the
Decasteljau algorithm at different times."""
ax = plt.subplot(1, 1, 1)
plt.subplots_adjust(bottom=0.25)
plt.axis([-1, 1, -1, 1])
axT = plt.axes([0.25, 0.1, 0.65, 0.03])
sT = wg.Slider(axT, 't', 0, 1, valinit=0)
pts = plt.ginput(n, timeout=240)
plt.subplot(1, 1, 1)
pts = np.array(pts)
plt.cla()
T0=0
decasteljau_animated(pts, T0)
plt.axis([-1, 1, -1, 1])
plt.draw()
def update(val):
T = sT.val
ax.cla()
plt.subplot(1, 1, 1)
decasteljau_animated(pts,T)
plt.xlim((-1, 1))
plt.ylim((-1,1))
plt.draw()
sT.on_changed(update)
plt.show()
def measure(self,line):
h,w = line.shape
smoothed = filters.gaussian_filter(line,(h*0.5,h*self.smoothness),mode='constant')
smoothed += 0.001*filters.uniform_filter(smoothed,(h*0.5,w),mode='constant')
self.shape = (h,w)
a = np.argmax(smoothed,axis=0)
a = filters.gaussian_filter(a,h*self.extra)
self.center = np.array(a,'i')
deltas = np.abs(np.arange(h)[:,np.newaxis]-self.center[np.newaxis,:])
self.mad = np.mean(deltas[line!=0])
self.r = int(1+self.range*self.mad)
if self.debug:
plt.figure("center")
plt.imshow(line,cmap=plt.cm.gray)
plt.plot(self.center)
plt.ginput(1,1000)
def infernoise_interactive(deg=4,lmbda=0.):
plt.axes()
plt.xlim([-10,10])
plt.ylim([-10,10])
a = PolyRegressionInferNoise(w_0=zeros(deg+1),V_0=eye(deg+1),a_0=5,b_0=1.)
t = np.linspace(-10,10,100)
points = []
while True:
newpoint = plt.ginput()
if newpoint == []:
return a
points += newpoint
x,y = points[-1]
plt.cla()
plt.plot([p[0] for p in points],[p[1] for p in points],'kx')
a.condition_on(y,x)
predictions, errors = a.predict(t)
plt.plot(t,predictions,'b-')
plt.plot(t,predictions+np.sqrt(errors),'b--')
plt.plot(t,predictions-np.sqrt(errors),'b--')
for i in range(3):
predictions, _ = a.predict_sample(t)
plt.plot(t,predictions,'r-',alpha=0.5)
plt.xlim([-10,10])
plt.ylim([-10,10])
def normalize_spectrum(wl, net, kernel_size = 201):
fig = pyplot.figure(figsize = [22, 7])
ax = fig.add_subplot(1,1,1)
wl = wl.flatten()
net = net.flatten()
ax.plot(wl, net, 'b')
background_wl = np.empty((0,))
background_net = np.empty((0,))
raw_input('Zoom in on spectrum and press enter to continue ')
select_another_region = 'y'
print 'Select background regions '
while select_another_region != 'n':
pt1, pt2 = pyplot.ginput(n = 2, timeout = -1)
x1, y1 = pt1
x2, y2 = pt2
background_indx = np.where((wl < max(x1, x2)) & (wl > min(x1, x2)))
background_wl = np.append(background_wl, wl[background_indx])
background_net = np.append(background_net, net[background_indx])
pyplot.plot([wl[background_indx[0][0]], wl[background_indx[0][-1]]], [net[background_indx[0][0]], net[background_indx[0][-1]]], 'm--|', lw = 3)
select_another_region = raw_input('Would you like to select another background regions? (y), n ')
background_wl = background_wl.flatten()
background_net = background_net.flatten()
sort_indx = np.argsort(background_wl)
background_wl = background_wl[sort_indx]
background_net = background_net[sort_indx]
poly_coeff = np.polyfit(background_wl, background_net, 3)
continuum = np.polyval(poly_coeff, wl)
ax.plot(wl, continuum, 'c', lw = 2)
pyplot.draw()
return wl, net/continuum, continuum
def get_shape(img, img_name, extension):
shape_filename = "data/" + img_name + "_points" + extension
if os.path.isfile(shape_filename):
shape = np.loadtxt(shape_filename)
else:
img_plot = plt.imshow(img)
prev_input, x_coor, y_coor = (0, 0), [], []
while True:
x, y = plt.ginput(1)[0]
x, y = int(x), int(y)
if (x, y) == prev_input: break
prev_input = (x, y)
x_coor.append(x); y_coor.append(y);
plt.plot(x_coor, y_coor, 'y')
plt.draw()
plt.close('all')
coordinates = np.array((np.array(x_coor), np.array(y_coor)))
shape = coordinates.T
tlc, trc, blc, brc = [0, 0], [0, img.shape[0]], [img.shape[1], 0], [img.shape[1], img.shape[0]]
shape = np.vstack([shape, np.array([tlc, trc, blc, brc])])
np.savetxt(shape_filename, shape)
return shape
def select_points(img, img_name, folder, extension):
shape_filename = "data/" + folder + img_name + "_points" + extension
if os.path.isfile(shape_filename):
shape = np.loadtxt(shape_filename)
else:
print ("Please select correspondence points for the image.")
img_plot = plt.imshow(img)
plt.suptitle('Please select points')
points, x_coor, y_coor = (0, 0), [], []
plt.xlim(0, img.shape[1])
plt.ylim(img.shape[0], 0)
while True:
x, y = plt.ginput(n=1, timeout=0)[0]
x, y = int(x), int(y)
if points and (x, y) == points: break;
points = (x, y)
x_coor.append(x); y_coor.append(y);
plt.plot(x_coor, y_coor, 'bs')
plt.draw()
plt.close('all')
coordinates = np.array((np.array(x_coor), np.array(y_coor)))
shape = coordinates.T
np.savetxt(shape_filename, shape)
return shape
from nilearn import plotting
a = np.array(skimage.transform.resize(img.dataobj, (60, 90)))
img_gray = a[:, :, 128]
#normalizo la imagen, el rango irá a partir de ahora [0,1], así que ojo a la hora de meter rangos en region growing
img_o = img_gray
img_o = img_o / np.max(img_o)
#%%
#def RegionGrowingP2(img, rango_inferior, rango_superior):
plt.imshow(img_o, cmap='gray')
click_markers = plt.ginput(n=1, timeout=30)
print(click_markers[0])
click_markers = list(click_markers[0])
print(click_markers)
markers = [round(num) for num in click_markers]
seed = [markers[1], markers[0]]
print(seed)
# [img_pad[i-1,j-1],img_pad[i-1,j],img_pad[i-1,j+1]],
# [img_pad[i,j-1],img_pad[i,j],img_pad[i,j+1]],
# [img_pad[i+1,j-1],img_pad[i+1,j],img_pad[i+1,j+1]
#%%
coords = np.array([(1, 0), (1, 1), (0, 1), (-1, 1), (-1, 0), (-1, -1), (0, -1),
(1, -1)])
#plt.subplot(1,2,2)
#plt.plot(y_axis,his1/his1.sum(), linewidth=3, alpha=0.5)
#plt.xlabel(r'$\varphi \ (Degrees)$', fontsize=25)
fig = plt.figure(figsize=(10, 8))
plt.contourf(-np.ma.log(his.T / his.sum()),
25,
cmap='gnuplot',
extent=model.extent)
plt.xlabel(r'$X$')
plt.ylabel(r'$Y$')
plt.colorbar()
fig.show()
coords = []
#def onclick(event):
# global ix, iy
# ix, iy = event.xdata, event.ydata
# print('x = %d, y = %d'%(
# ix, iy))
#
# global coords
# coords.append((ix, iy))
# return coords
#cid = fig.canvas.mpl_connect('button_press_event', onclick)
coords = plt.ginput(-1, show_clicks=True)
#plt.savefig("fes_"+str(mode)+".pdf")
print(coords)
np.savetxt(path + 'coords_' + str(mode) + '.dat', coords)
def getCorrespondence(image1, image2, numOfPoints_=8, manualSelection_=True):
manualSelection = manualSelection_
numOfPoints = numOfPoints_
if (numOfPoints < 8):
print("Error: Num of paris must be greater or equal 4")
return
if (manualSelection):
# Display images, select matching points
fig = plt.figure()
fig1 = fig.add_subplot(1, 2, 1)
fig2 = fig.add_subplot(1, 2, 2)
# Display the image
fig1.imshow(image1)
fig2.imshow(image2)
plt.axis('image')
pts = plt.ginput(n=numOfPoints, timeout=0)
pts = np.reshape(pts, (2, int(numOfPoints / 2), 2))
# print(pts);
return pts
else:
"""
automatic selection process
"""
numOfPoints = 20
img1 = cv2.imread('../images/mountain/image1.jpg', 0)
img2 = cv2.imread('../images/mountain/image2.jpg', 0)
# Initiate SIFT detector
orb = cv2.ORB()
# find the keypoints and descriptors with SIFT
kp1, des1 = orb.detectAndCompute(img1, None)
kp2, des2 = orb.detectAndCompute(img2, None)
# create BFMatcher object
bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)
# Match descriptors.
matches = bf.match(des1, des2)
# Sort them in the order of their distance.
matches = sorted(matches, key=lambda x: x.distance)
print(matches)
return matches
# 10 to 10 mountain points
pts = np.array([(290.99673461310101, 168.49247971502086),
(351.77211557490591, 132.47743914506236),
(381.03433603799715, 139.23025925192962),
(455.31535721353646, 161.7396596081536),
(471.07193746289317, 148.23401939441919),
(511.58885810409652, 175.24529982188801),
(507.08697803285168, 359.82238274292507),
(403.54373639422113, 319.30546210172179),
(390.03809618048672, 341.81486245794582),
(290.99673461310101, 359.82238274292507),
(5.9867999208391893, 139.23025925192951),
(80.267821096378498, 109.96803878883827),
(107.27910152384732, 121.22273896695026),
(177.05824262814178, 161.73965960815349),
(195.06576291312103, 148.23401939441908),
(224.32798337621227, 175.2452998218879),
(208.57140312685544, 353.06956263605764),
(98.275341381357748, 303.54888185236484),
(82.518761132000918, 326.05828220858882),
(3.7358598852168825, 353.06956263605764)])
pts = np.reshape(pts, (2, int(numOfPoints / 2), 2))
return pts
def womgau(hop):
"""fit gaussian to line"""
import numpy as np
import logging
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit
from tmath.wombat.womwaverange import womwaverange
from tmath.wombat.womget_element import womget_element
from tmath.wombat.inputter import inputter
from tmath.wombat.inputter_single import inputter_single
from tmath.wombat.gauss import gauss
from tmath.wombat.gauss_cont import gauss_cont
from tmath.wombat.yesno import yesno
print(' ')
logging.info('Object is {}'.format(hop[0].obname))
print(' ')
print('Spectrum runs from {} to {}'.format(hop[0].wave[0],
hop[0].wave[-1]))
print(' ')
print('This routine expects the spectrum to be in flambda units.')
print('It also expects a linear wavelength scale.')
print(' ')
print('Choose general region of spectrum\n')
nwave, nflux, mode = womwaverange(hop[0].wave, hop[0].flux, 'none')
print('\nNow pick the exact range for the fit')
waveint, fluxint, mode = womwaverange(nwave, nflux, mode)
indexblue = womget_element(nwave, waveint[0])
indexred = womget_element(nwave, waveint[-1])
if (mode == 'w'):
done = False
while (not done):
print(' ')
wavecenter = inputter('Enter approximate center of Gaussian : ',
'float', False)
indexcenter = womget_element(waveint, wavecenter)
if (indexcenter <= 0) or (wavecenter > waveint[-1]):
print('Bad central wavelength, try again')
else:
done = True
else:
done = False
while (not done):
print('Mark the approximate center of the Gaussian')
pickcent = plt.ginput(1, timeout=-1)
indexcenter = womget_element(waveint, pickcent[0][0])
print('\nApproximate center at {}'.format(waveint[indexcenter]))
print('\nIs this OK?')
answer = yesno('y')
if (answer == 'y'):
done = True
weights = np.sqrt(hop[0].var[indexblue:indexred + 1])
print(' ')
continuum = inputter_single(
'Do you want to fit gaussian with (c)ontinuum, or (n)o continuum? ',
'cn')
if (continuum == 'c'):
p = [fluxint[indexcenter], waveint[indexcenter], 3.0, 1.0, waveint[0]]
result = curve_fit(gauss_cont,
waveint,
fluxint,
sigma=weights,
p0=p,
absolute_sigma=True,
full_output=True)
else:
p = [fluxint[indexcenter], waveint[indexcenter], 3.0]
result = curve_fit(gauss,
waveint,
fluxint,
sigma=weights,
p0=p,
absolute_sigma=True,
full_output=True)
coefferr = np.sqrt(np.diag(result[1]))
coeff = result[0]
# make 'finer-grained' version of fit, 0.2A/pix for calculations
wavecalc = np.arange(2 * 5 * 50 * abs(
coeff[2])) * 0.2 + coeff[1] - 0.2 * 5 * 50 * abs(coeff[2])
calccenter = womget_element(wavecalc, coeff[1])
if (continuum == 'c'):
fluxcalc = gauss_cont(wavecalc, *coeff)
fluxcont = wavecalc * coeff[3] + coeff[4]
fluxgaussian = fluxcalc - fluxcont
linecont = fluxcont[calccenter]
else:
fluxcalc = gauss(wavecalc, *coeff)
deltafit = wavecalc[1] - wavecalc[0]
calcindexblue = womget_element(wavecalc, waveint[0])
calcindexred = womget_element(wavecalc, waveint[-1])
sumfluxcalc = np.sum(fluxcalc[calcindexblue:calcindexred + 1] * deltafit)
sumallfluxcalc = np.sum(fluxcalc * deltafit)
chi = (result[2]['fvec']**2).sum()
redchi = chi / (len(waveint) - len(coeff))
if (continuum == 'c'):
sumfluxgaussian = np.sum(fluxgaussian[calcindexblue:calcindexred + 1] *
deltafit)
sumallfluxgaussian = np.sum(fluxgaussian * deltafit)
sumfluxcont = np.sum(fluxcont[calcindexblue:calcindexred + 1] *
deltafit)
sumallfluxcont = np.sum(fluxcont * deltafit)
# propagate uncertainty (from old version) not sure this is correct
height_pct = coefferr[0] / coeff[0]
sigma_pct = coefferr[2] / coeff[2]
flux_pct = np.sqrt(height_pct**2 + sigma_pct**2)
sumfluxgaussiansig = sumfluxgaussian * flux_pct
sumallfluxgaussiansig = sumallfluxgaussian * flux_pct
plt.cla()
plt.plot(nwave, nflux, drawstyle='steps-mid', color='k')
plt.ylabel('Flux')
plt.xlabel('Wavelength')
xmin, xmax = plt.xlim()
ymin, ymax = plt.ylim()
plt.plot(wavecalc, fluxcalc, drawstyle='steps-mid', color='b')
if (continuum == 'c'):
plt.plot(wavecalc, fluxgaussian, drawstyle='steps-mid', color='r')
plt.plot(wavecalc, fluxcont, drawstyle='steps-mid', color='g')
plt.plot([waveint[0], waveint[0]], [ymin, ymax], color='k', linestyle='--')
plt.plot([waveint[-1], waveint[-1]], [ymin, ymax],
color='k',
linestyle='--')
plt.xlim([xmin, xmax])
plt.ylim([ymin, ymax])
logging.info('For object {} Gaussian fit'.format(hop[0].obname))
if (continuum == 'c'):
print(
'\nData = Black, Fit = Blue, Continuum = Green, Fit-Continuum = Red\n'
)
else:
print('\nData = Black, Fit = Blue\n')
logging.info('Height {:16.8f}+/-{:16.8f}'.format(
coeff[0], coefferr[0]))
logging.info('Center {:16.8f}+/-{:16.8f}'.format(
coeff[1], coefferr[1]))
logging.info('Sigma {:16.8f}+/-{:16.8f}'.format(
coeff[2], coefferr[2]))
if (continuum == 'c'):
logging.info('Slope {:16.8f}+/-{:16.8f}'.format(
coeff[3], coefferr[3]))
logging.info('Y-intercept {:16.8f}+/-{:16.8f}'.format(
coeff[4], coefferr[4]))
logging.info('FWHM {:16.8f}+/-{:16.8f}'.format(
2.35482 * np.abs(coeff[2]), 2.35482 * coefferr[2]))
logging.info(
'Flux between dotted lines (Gaussian): {:16.8f}+/-{:16.8f}'.format(
sumfluxgaussian, sumfluxgaussiansig))
logging.info('EW between dotted lines (Gaussian): {:16.8f}'.format(
sumfluxgaussian / linecont))
logging.info('Flux for full (Gaussian): {:16.8f}+/-{:16.8f}'.format(
sumallfluxgaussian, sumallfluxgaussiansig))
logging.info('EW for full (Gaussian): {:16.8f}'.format(
sumallfluxgaussian / linecont))
logging.info(
'Continuum flux at line center: {:16.8f}'.format(linecont))
logging.info('Chi^2: {}'.format(chi))
logging.info('Reduced chi^2: {}'.format(redchi))
logging.info('All fluxes might need to be scaled by 1e-15')
print(' ')
return hop
def normalize(self, plot=True, norm_method='linear'):
"""
Normalize the region if the data are not already normalized.
Choose from two methods:
1: define left and right continuum regions
and fit a linear continuum.
2: define the continuum as a range of points
and use spline interpolation to infer the
continuum.
"""
if norm_method == 'linear':
norm_num = 1
elif norm_method == 'spline':
norm_num = 2
else:
err_msg = "Invalid norm_method: %r" % norm_method
raise ValueError(err_msg)
plt.close('all')
plt.figure()
dx = 0.1 * (self.wl.max() - self.wl.min())
lines_title_string = ", ".join([line.tag for line in self.lines])
plt.xlim(self.wl.min() - dx, self.wl.max() + dx)
plt.ylim(0.8 * self.flux.min(), 1.2 * self.flux.max())
plt.plot(self.wl,
self.flux,
color='k',
drawstyle='steps-mid',
label=lines_title_string)
plt.xlabel("Wavelength [${\\rm \AA}$]")
if norm_num == 1:
# - Normalize by defining a left and right continuum region
print "\n\n Mark left continuum region, left and right boundary."
plt.title("Mark left continuum region, left and right boundary.")
bounds = plt.ginput(2, -1)
left_bound = min(bounds[0][0], bounds[1][0])
right_bound = max(bounds[0][0], bounds[1][0])
region1 = (self.wl >= left_bound) * (self.wl <= right_bound)
fit_wl = self.wl[region1]
fit_flux = self.flux[region1]
lines_title_string = ", ".join([line.tag for line in self.lines])
plt.title(lines_title_string)
print "\n Mark right continuum region, left and right boundary."
plt.title("Mark right continuum region, left and right boundary.")
bounds = plt.ginput(2)
left_bound = min(bounds[0][0], bounds[1][0])
right_bound = max(bounds[0][0], bounds[1][0])
region2 = (self.wl >= left_bound) * (self.wl <= right_bound)
fit_wl = np.concatenate([fit_wl, self.wl[region2]])
fit_flux = np.concatenate([fit_flux, self.flux[region2]])
popt, pcov = curve_fit(linfunc, fit_wl, fit_flux)
continuum = linfunc(self.wl, *popt)
e_continuum = np.std(fit_flux - linfunc(fit_wl, *popt))
elif norm_num == 2:
# Normalize by drawing the continuum and perform spline
# interpolation between the points
print "\n\n Select a range of continuum spline points over the whole range"
plt.title(
" Select a range of continuum spline points over the whole range"
)
points = plt.ginput(n=-1, timeout=-1)
xk, yk = [], []
for x, y in points:
xk.append(x)
yk.append(y)
xk = np.array(xk)
yk = np.array(yk)
region_wl = self.wl.copy()
continuum = spline(xk, yk, region_wl, order=3)
e_continuum = np.sqrt(np.mean(self.err**2))
if plot:
new_flux = self.flux / continuum
new_err = self.err / continuum
plt.cla()
plt.plot(self.wl,
new_flux,
color='k',
drawstyle='steps-mid',
label=lines_title_string)
plt.xlabel("Wavelength [${\\rm \AA}$]")
plt.title("Normalized")
plt.axhline(1., ls='--', color='k')
plt.axhline(1. + e_continuum / np.mean(continuum),
ls=':',
color='gray')
plt.axhline(1. - e_continuum / np.mean(continuum),
ls=':',
color='gray')
plt.show(block=False)
plt.title("Go back to terminal...")
prompt = raw_input(" Is normalization correct? (YES/no) ")
if prompt.lower() in ['', 'y', 'yes']:
self.flux = new_flux
self.err = new_err
self.cont_err = e_continuum / np.median(continuum)
self.normalized = True
return 1
else:
return 0
else:
self.flux = self.flux / continuum
self.err = self.err / continuum
self.cont_err = e_continuum / np.mean(continuum)
self.normalized = True
return 1
def run_align_atlas_gui_simple(img, atlas, fig_path):
"""
This is DEPRECATED code as of 20180709, replaced by run_align_atlas_gui.
Potentially to be deleted eventually.
To add:
Simple scaling and translation, based on the original
two keypoints (which set the orientation).
With real time update/feedback.
:param img:
:param atlas:
:param fig_path:
:return:
"""
warning('run_align_atlas_gui_simple is DEPRECATED.')
keypoint_positions = ['Anterior midline', 'Posterior midline']
# Load atlas.
atlas, annotations, atlas_outline = load_atlas()
do_manual_atlas_keypoint = False
if do_manual_atlas_keypoint:
atlas_coords = []
plt.figure(figsize=(30, 30))
plt.imshow(atlas_outline)
for t in keypoint_positions:
plt.title('Click on ' + t)
c = plt.ginput(n=1)
plt.plot(c[0][0], c[0][1], 'ro')
atlas_coords.extend(c)
else:
atlas_coords = [(98, 227), (348, 227)]
plt.figure(figsize=(30, 30))
plt.imshow(atlas_outline)
for cc in atlas_coords:
plt.plot(cc[0], cc[1], 'ro')
# Convert selected keypoints to array.
atlas_coords_array = np.zeros((len(atlas_coords), 2))
for ci, c in enumerate(atlas_coords):
atlas_coords_array[ci, 0] = np.round(c[0])
atlas_coords_array[ci, 1] = np.round(c[1])
while 1:
plt.figure()
plt.imshow(img)
img_coords = []
do_manual_patch_keypoint = True
if do_manual_patch_keypoint:
plt.figure(figsize=(30, 30))
plt.imshow(img)
for t in keypoint_positions:
plt.title('Click on ' + t)
c = plt.ginput(n=1)
plt.plot(c[0][0], c[0][1], 'ro')
img_coords.extend(c)
else:
# These numbers are just for fast debugging.
img_coords = [(26, 297), (430, 314)]
plt.figure(figsize=(30, 30))
plt.imshow(img, cmap='Greys_r')
for cc in img_coords:
plt.plot(cc[0], cc[1], 'ro')
plt.close('all')
# Convert selected keypoints to array.
img_coords_array = np.zeros((len(img_coords), 2))
for ci, c in enumerate(img_coords):
img_coords_array[ci, 0] = np.round(c[0])
img_coords_array[ci, 1] = np.round(c[1])
aligned_atlas_outline, aligned_img, tform = align_atlas_to_image(
atlas_outline,
img,
atlas_coords_array[0:2, :],
img_coords_array[0:2, :],
do_debug=False)
# Overlay atlas on image for checking that things look good.
overlay = overlay_atlas_outline(aligned_atlas_outline, img)
plt.figure(figsize=(20, 20))
plt.imshow(overlay, cmap='Greys_r')
plt.title(
'Check that things look good, and close this window manually.')
plt.show()
text = input('Look good? [y] or [n]')
print(text)
if text == 'y':
break
# # Save out selections
# keypoints_dir = os.path.join(self._keypoints_file, name, \
# str(name) + '_source_extraction')
# save_fname = os.path.join(keypoints_dir, 'keypoints.npz')
# print('Saving keypoints and aligned atlas to: ' + save_fname)
# np.savez(save_fname,
# coords=img_coords_array,
# atlas_coords=atlas_coords,
# atlas=atlas,
# img=aligned_img,
# aligned_atlas_outline=aligned_atlas_outline)
plt.figure()
plt.imshow(img, cmap='Greys_r')
for ci, c in enumerate(img_coords):
print(c)
plt.plot(c[0], c[1], 'ro')
plt.savefig(fig_path + '_keypoints.png')
plt.figure()
plt.imshow(overlay, cmap='Greys_r')
plt.savefig(fig_path + '_overlay.png')
return (atlas_coords_array, img_coords_array, aligned_atlas_outline, atlas)
# Create sampling space
#x = np.linspace(0, 100, 10)
#y = np.linspace(0, 80, 10)
#X, Y = np.meshgrid(x, y)
#print(X.shape)
#T_lst = [gate_lst[i].samplePoints(X, Y, object_lst) for i in range(len(gate_lst))]
#ax.pcolormesh(X, Y, T_lst[0]+T_lst[1]+T_lst[2])
ax.set_title('City cross')
ax.set_xlabel('x')
ax.set_ylabel('y')
plt.xlim([0, 100])
plt.ylim([0, 80])
sampled_pts = plt.ginput(200, timeout=0)
sampled_pts = np.array([[p[0], p[1]] for p in sampled_pts])
samp_x = sampled_pts[:, 0]
samp_y = sampled_pts[:, 1]
T0 = np.array([gate_lst[0].samplePoint(sampled_pts[i,:], object_lst) for i in range(samp_x.shape[0])])
T1 = np.array([gate_lst[1].samplePoint(sampled_pts[i,:], object_lst) for i in range(samp_x.shape[0])])
T2 = np.array([gate_lst[2].samplePoint(sampled_pts[i,:], object_lst) for i in range(samp_x.shape[0])])
T = np.stack((T0, T1, T2), axis=1)
T_color = T / 100.
#T = normalize(T, axis=1)
ax.scatter(samp_x, samp_y, c=T_color, edgecolor='none', s=np.max(T_color, axis=1)*150 )
plt.pause(0.1)
def select_keypoints(atlas_outline, img):
"""
Presents an interface for selecting two
keypoints (along the midline) that define
the initial position (and importantly, the
rotation) of the aligned atlas.
:param atlas_outline: Outline to overlay on atlas.
:param img: Image to which the atlas will be aligned.
:return: atlas_coords_array: the keypoints in atlas space
img_coords_array: the corresponding keypoints in image space
(which were user selected).
"""
keypoint_positions = ['Anterior midline', 'Posterior midline']
do_manual_atlas_keypoint = False
if do_manual_atlas_keypoint:
atlas_coords = []
plt.figure(figsize=(30, 30))
plt.imshow(atlas_outline)
for t in keypoint_positions:
plt.title('Click on ' + t)
c = plt.ginput(n=1)
plt.plot(c[0][0], c[0][1], 'ro')
atlas_coords.extend(c)
else:
# These numbers were taken from an initial manual selection.
atlas_coords = [(98, 227), (348, 227)]
plt.figure(figsize=(30, 30))
plt.imshow(atlas_outline)
for cc in atlas_coords:
plt.plot(cc[0], cc[1], 'ro')
# Convert selected keypoints to array.
atlas_coords_array = np.zeros((len(atlas_coords), 2))
for ci, c in enumerate(atlas_coords):
atlas_coords_array[ci, 0] = np.round(c[0])
atlas_coords_array[ci, 1] = np.round(c[1])
while 1:
plt.figure()
plt.imshow(img)
img_coords = []
do_manual_patch_keypoint = True
if do_manual_patch_keypoint:
plt.figure(figsize=(30, 30))
plt.imshow(img)
for t in keypoint_positions:
plt.title('Click on ' + t)
c = plt.ginput(n=1)
plt.plot(c[0][0], c[0][1], 'ro')
img_coords.extend(c)
else:
# These numbers are just for fast debugging.
img_coords = [(26, 297), (430, 314)]
# (26, 187),
# (17, 420)]
plt.figure(figsize=(30, 30))
plt.imshow(img, cmap='Greys_r')
for cc in img_coords:
plt.plot(cc[0], cc[1], 'ro')
plt.close('all')
# Convert selected keypoints to array.
img_coords_array = np.zeros((len(img_coords), 2))
for ci, c in enumerate(img_coords):
img_coords_array[ci, 0] = np.round(c[0])
img_coords_array[ci, 1] = np.round(c[1])
return atlas_coords_array, img_coords_array
# *************************************************
numCones = rick.load(open("numCones.pkl", "rb"))
numNCones = rick.load(open("numNCones.pkl", "rb"))
color_histogram = []
if ((getNewTrainingCones) or (getNewTrainingNCones)):
plot.figure(plotIndex)
plotIndex += 1
plot.clf()
plot.imshow(im)
if initializeNewLists:
trainingImages = []
labels = []
if getNewTrainingCones:
for index in range(newConesToAddToTrainingImage):
p = plot.ginput(2)
points = np.array([[p[0][0], p[0][1]], [p[1][0], p[1][1]]])
cone = cv2.resize(
im[int(min(points[:, 1])):int(max(points[:, 1])),
int(min(points[:, 0])):int(max(points[:, 0]))], (res, res))
trainingImages.append(cone)
labels.append(1)
numCones += 1
color_histogram.append(
np.histogram(cone, 50)[0]
) # Color Histogram (Not actually used for anything because it gave trash results for me. My other methods worked better)
plot.imsave(('ConeImages/coneImage' + str(numCones) + '.bmp'),
cone)
if getNewTrainingNCones:
for index in range(newNConesToAddToTrainingImage):
p = plot.ginput(2)
def define_mask(self, z=None, dataset=None, telluric=True):
"""
Use an interactive window to define the mask for the region.
Parameters
----------
z : float [default = None]
If a redshift is given, the lines in the region are shown as vertical lines
at the given redshift.
dataset : :class:`VoigtFit.DataSet` [default = None]
A dataset with components defined for the lines in the region.
If a dataset is passed, the components of the lines in the region are shown.
telluric : bool [default = True]
Show telluric absorption and sky emission line templates during the masking.
"""
plt.close('all')
plt.xlim(self.wl.min(), self.wl.max())
# plt.ylim(max(0, 0.8*self.flux.min()), 1.2)
lines_title = ", ".join([line.tag for line in self.lines])
plt.plot(self.wl,
self.flux,
color='k',
drawstyle='steps-mid',
lw=0.5,
label=lines_title)
plt.xlabel("Wavelength [${\\rm \AA}$]")
plt.legend()
if telluric:
wl_T = telluric_data['wl']
cutout = (wl_T > self.wl.min()) * (wl_T < self.wl.max())
flux_T = telluric_data['em'][cutout]
abs_T = telluric_data['abs'][cutout]
wl_T = wl_T[cutout]
if self.normalized:
cont = 1.
else:
cont = np.median(self.flux)
plt.plot(wl_T,
abs_T * 1.2 * cont,
color='crimson',
alpha=0.7,
lw=0.5)
plt.plot(wl_T, (flux_T / flux_T.max() + 1.2) * cont,
color='orange',
alpha=0.7,
lw=0.5)
if z is not None:
for line in self.lines:
# Load line properties
l0, f, gam = line.get_properties()
if dataset is not None:
ion = line.ion
n_comp = len(dataset.components[ion])
ion = ion.replace('*', 'x')
for n in range(n_comp):
z = dataset.pars['z%i_%s' % (n, ion)].value
plt.axvline(l0 * (z + 1), ls=':', color='r', lw=0.4)
else:
plt.axvline(l0 * (z + 1), ls=':', color='r', lw=0.4)
plt.title("Mark regions to mask, left and right boundary.")
print "\n\n Mark regions to mask, left and right boundary."
plt.draw()
ok = 0
mask_vlines = list()
while ok >= 0:
sel = plt.ginput(0, timeout=-1)
if len(sel) > 0 and len(sel) % 2 == 0:
mask = self.mask.copy()
sel = np.array(sel)
selections = np.column_stack([sel[::2, 0], sel[1::2, 0]])
for x1, x2 in selections:
cutout = (self.wl >= x1) * (self.wl <= x2)
mask[cutout] = False
mask_vlines.append(plt.axvline(x1, color='r', ls='--'))
mask_vlines.append(plt.axvline(x2, color='r', ls='--'))
masked_spectrum = np.ma.masked_where(mask, self.flux)
mask_line = plt.plot(self.wl,
masked_spectrum,
color='r',
drawstyle='steps-mid')
plt.draw()
prompt = raw_input(
"Are the masked regions correct? (YES/no/clear)")
if prompt.lower() in ['', 'y', 'yes']:
ok = -1
self.mask = mask
self.new_mask = False
elif prompt.lower() in ['c', 'clear']:
ok = 0
self.mask = np.ones_like(mask, dtype=bool)
for linesegment in mask_line:
linesegment.remove()
mask_line = list()
for linesegment in mask_vlines:
linesegment.remove()
mask_vlines = list()
else:
self.mask = mask
ok += 1
elif len(sel) == 0:
print "\nNo masks were defined."
prompt = raw_input("Continue? (yes/no)")
if prompt.lower() in ['', 'y', 'yes']:
ok = -1
self.new_mask = False
else:
ok += 1
slice_path = sub_all_tif[i - 1]
slice_img = cv2.imread(slice_path, -1)
# now get the slice image
masked_image = add_mask(mask_centre, radius, slice_img)
x_coordinate, y_coordinate = random_effective_area(masked_image)
plt.imshow(
masked_image[x_coordinate - show_length:x_coordinate + show_length,
y_coordinate - show_length:y_coordinate +
show_length], 'gray')
plt.title(
'Please label any points for pore! ({:d}/{:d}) \n Current slice: {str}'
.format((index + 1), num_slices, str=os.path.basename(slice_path)),
color='red')
coordinate = plt.ginput(n=num_points, timeout=0)
# note that x, y from the ginput function is oppisite to our x and y, we need to transfer it
transformed_coordinate = transform(coordinate, x_coordinate,
y_coordinate, show_length)
print(transformed_coordinate)
with open(file_path, 'a') as f:
f.writelines(
[slice_path, ' ',
str(transformed_coordinate), ' ', '0', '\n'])
# '0' means pore
print(
'Please label any points for non-pore (not artifact) in each picture!')
# Then we label all non-pore
def normalize(self, plot=True, norm_method='linear', z_sys=None):
"""
Normalize the region if the data are not already normalized.
Choose from two methods:
1: define left and right continuum regions
and fit a linear continuum.
2: define the continuum as a range of points
and use spline interpolation to infer the
continuum.
If `z_sys` is not `None`, show the region in velocity space using
instead of wavelength space.
"""
if norm_method in ['linear', 'spline']:
pass
else:
err_msg = "Invalid norm_method: %r" % norm_method
raise ValueError(err_msg)
plt.close('all')
plt.figure()
x = self.wl.copy()
x_label = u"Wavelength [Å]"
if z_sys is not None:
# Calculate velocity:
l0 = self.lines[0].l0 * (z_sys + 1.)
x = (x - l0) / l0 * 299792.458
x_label = u"Rel. Velocity [${\\rm km\\ s^{-1}}$]"
dx = 0.1 * (x.max() - x.min())
lines_title_string = ", ".join([line.tag for line in self.lines])
plt.xlim(x.min() - dx, x.max() + dx)
plt.ylim(0.8 * self.flux.min(), 1.2 * self.flux.max())
plt.plot(x,
self.flux,
color='k',
drawstyle='steps-mid',
label=lines_title_string)
plt.xlabel(x_label)
if norm_method == 'linear':
# - Normalize by defining a left and right continuum region
print "\n\n Mark left continuum region, left and right boundary."
plt.title("Mark left continuum region, left and right boundary.")
bounds = plt.ginput(2, -1)
left_bound = min(bounds[0][0], bounds[1][0])
right_bound = max(bounds[0][0], bounds[1][0])
region1 = (x >= left_bound) * (x <= right_bound)
fit_wl = x[region1]
fit_flux = self.flux[region1]
lines_title_string = ", ".join([line.tag for line in self.lines])
plt.title(lines_title_string)
print "\n Mark right continuum region, left and right boundary."
plt.title("Mark right continuum region, left and right boundary.")
bounds = plt.ginput(2)
left_bound = min(bounds[0][0], bounds[1][0])
right_bound = max(bounds[0][0], bounds[1][0])
region2 = (x >= left_bound) * (x <= right_bound)
fit_wl = np.concatenate([fit_wl, x[region2]])
fit_flux = np.concatenate([fit_flux, self.flux[region2]])
popt, pcov = curve_fit(linfunc, fit_wl, fit_flux)
continuum = linfunc(x, *popt)
e_continuum = np.std(fit_flux - linfunc(fit_wl, *popt))
elif norm_method == 'spline':
# Normalize by drawing the continuum and perform spline
# interpolation between the points
print "\n\n Select a range of continuum spline points over the whole range"
plt.title(
" Select a range of continuum spline points over the whole range"
)
points = plt.ginput(n=-1, timeout=-1)
points = np.array(points)
xk = points[:, 0]
yk = points[:, 1]
# region_wl = self.wl.copy()
cont_spline = spline(xk, yk, s=0.)
continuum = cont_spline(x)
e_continuum = np.sqrt(np.mean(self.err**2))
if plot:
new_flux = self.flux / continuum
new_err = self.err / continuum
plt.cla()
plt.plot(x,
new_flux,
color='k',
drawstyle='steps-mid',
label=lines_title_string)
plt.xlabel(x_label)
plt.title("Normalized")
plt.axhline(1., ls='--', color='k')
plt.axhline(1. + e_continuum / np.mean(continuum),
ls=':',
color='gray')
plt.axhline(1. - e_continuum / np.mean(continuum),
ls=':',
color='gray')
plt.draw()
plt.title("Go back to terminal...")
prompt = raw_input(" Is normalization correct? (YES/no) ")
if prompt.lower() in ['', 'y', 'yes']:
self.flux = new_flux
self.err = new_err
self.cont_err = e_continuum / np.median(continuum)
self.normalized = True
return 1
else:
return 0
else:
self.flux = self.flux / continuum
self.err = self.err / continuum
self.cont_err = e_continuum / np.mean(continuum)
self.normalized = True
return 1
newedge.append((s, t))
edge = newedge
print(i)
if __name__ == '__main__':
# img_path = 'rect-test-1000.jpg'
img_path = 'maze.bmp'
# first mine
img0 = Image.open(img_path).convert('RGB')
t1 = time.time()
flood_fill_mine(img0, (1, 1), (0, 255, 0), None, 0)
print(time.time() - t1)
plt.imshow(img0)
plt.ginput(1)
plt.close()
# first mine
img0 = Image.open(img_path).convert('RGB')
t1 = time.time()
flood_fill_0(img0, (1, 1), (0, 255, 0), 0)
print(time.time() - t1)
plt.imshow(img0)
plt.ginput(1)
plt.close()
# second original
img1 = Image.open(img_path).convert('RGB')
t1 = time.time()
floodfill_o(img1, (1, 1), (0, 255, 0), None, 0)
conjunto_normalizado = normalizar(atributos)
plt.scatter(conjunto_normalizado[:, 0], conjunto_normalizado[:, 1])
plt.show()
# In[8]
# column_stack para aderir las clases al conjunto
conjunto_normalizado = np.column_stack((conjunto_normalizado, clases))
# Mostrar los valores pertenecientes
for i in range(-1, 2):
posicion, = np.where(conjunto_normalizado[:, 2] == i)
pertenecientes = conjunto_normalizado[posicion, :]
plt.scatter(pertenecientes[:, 0], pertenecientes[:, 1], 10)
# In[9]
# Obtener la nueva observacion para calcular sus vecinos mas cercanos
nueva_obs = plt.ginput(1)
observacion_array = np.tile(nueva_obs[0], (ren, 1))
plt.scatter(observacion_array[0, 0], observacion_array[0, 1], 30)
distEuc = distanciaEuclidiana(conjunto_normalizado[:, :col - 1],
observacion_array)
posicion = np.argsort(distEuc[:, 0])
# In[10]
# Plotear la nueva observacion con sus vecinos mas cercanos
for i in range(0, K):
plt.plot([observacion_array[i, 0], conjunto_normalizado[posicion[i], 0]],
[observacion_array[i, 1], conjunto_normalizado[posicion[i], 1]])
plt.show()
def check_E(img1, file1, img2, file2, E, K1, K2):
# input:
# img1 - a h x w x 3 numpy ndarray (dtype = np.unit8) holding the color
# image 1 (h, w being the height and width of the image)
# file1 - the filename of the color image 1
# img2 - a h x w x 3 numpy ndarray (dtype = np.unit8) holding the color
# image 2 (h, w being the height and width of the image)
# filen2 - the filename of the color image 2
# E - a 3 x 3 numpy ndarray holding the essential matrix
# K1 - a 3 x 3 numpy ndarray holding the K matrix of camera 1
# K2 - a 3 x 3 numpy ndarray holding the K matrix of camera 2
# form the fundamental matrix
F = inv(K2.T) @ E @ inv(K1)
plt.ion()
# plot image 1
fig1 = plt.figure('Epipolar geometry - {}'.format(file1))
plt.imshow(img1)
plt.autoscale(enable=False, axis='both')
# plot image 2
fig2 = plt.figure('Epipolar geometry - {}'.format(file2))
plt.imshow(img2)
plt.autoscale(enable=False, axis='both')
plt.show()
# ask user to pick points on image 1
print(
'please select points on {} using left-click, end with right-click...'.
format(file1))
plt.figure(fig1.number)
pt = plt.ginput(n=1, mouse_pop=2, mouse_stop=3, timeout=-1)
while pt:
pt = np.array(pt)
plt.figure(fig1.number)
plt.plot(pt[0, 0], pt[0, 1], 'rx')
# form the epipolar line
(a, b, c) = F @ np.array((pt[0, 0], pt[0, 1], 1))
(h, w, d) = img2.shape
x = np.array([0, w - 1])
y = np.array([(a * x[0] + c) / (-b), (a * x[1] + c) / (-b)])
if y[0] < 0 or y[0] > h - 1 or y[1] < 0 or y[1] > h - 1:
y = np.array([0, h - 1])
x = np.array([(b * y[0] + c) / (-a), (b * y[1] + c) / (-a)])
plt.figure(fig2.number)
plt.plot(x, y, 'r-')
plt.figure(fig1.number)
pt = plt.ginput(n=1, mouse_pop=2, mouse_stop=3, timeout=-1)
# ask user to pick points on image 1
print(
'please select points on {} using left-click, end with right-click...'.
format(file2))
plt.figure(fig2.number)
pt = plt.ginput(n=1, mouse_pop=2, mouse_stop=3, timeout=-1)
while pt:
pt = np.array(pt)
plt.figure(fig2.number)
plt.plot(pt[0, 0], pt[0, 1], 'bx')
# form the epipolar line
(a, b, c) = F.T @ np.array((pt[0, 0], pt[0, 1], 1))
(h, w, d) = img1.shape
x = np.array([0, w - 1])
y = np.array([(a * x[0] + c) / (-b), (a * x[1] + c) / (-b)])
if y[0] < 0 or y[0] > h - 1 or y[1] < 0 or y[1] > h - 1:
y = np.array([0, h - 1])
x = np.array([(b * y[0] + c) / (-a), (b * y[1] + c) / (-a)])
plt.figure(fig1.number)
plt.plot(x, y, 'b-')
plt.figure(fig2.number)
pt = plt.ginput(n=1, mouse_pop=2, mouse_stop=3, timeout=-1)
plt.close()
def run(self, image_filename_or_data, mask_color, opacity):
plt.ion()
plt.axis('off')
if isinstance(image_filename_or_data, str):
image_data = np.array(Image.open(image_filename_or_data))
elif isinstance(image_filename_or_data, list) or isinstance(image_filename_or_data, np.ndarray):
image_data = image_filename_or_data
else:
print("image_filename_or_data is error")
return
plt.imshow(image_data)
plt.title('Click one point of the object that you interested')
try:
while 1:
object_point = np.array(plt.ginput(1, timeout=0)).astype(np.int)[0]
where = [int(self.input_size[0] * object_point[1] / len(image_data)),
int(self.input_size[1] * object_point[0] / len(image_data[0]))]
print("point=[{},{}] where=[{},{}]".format(object_point[0], object_point[1], where[0], where[1]))
final_batch_data, data_raw, gaussian_mask = Data.load_image(image_data, where=where,
image_size=self.input_size)
print("begin to run ...")
# 运行
predict_output_r, pred_classes_r = self.sess.run([self.predict_output, self.pred_classes],
feed_dict={self.img_placeholder: final_batch_data})
print("end run")
# 类别
print("the class is {}({})".format(pred_classes_r[0], CategoryNames[pred_classes_r[0]]))
# 分割
segment = np.squeeze(np.asarray(np.where(predict_output_r[0] == 1, 1, 0), dtype=np.uint8))
segment = np.asarray(Image.fromarray(segment).resize((len(image_data[0]), len(image_data))))
image_mask = np.ndarray(image_data.shape)
image_mask[:, :, 0] = (1 - segment) * image_data[:, :, 0] + segment * (
opacity * mask_color[0] + (1 - opacity) * image_data[:, :, 0])
image_mask[:, :, 1] = (1 - segment) * image_data[:, :, 1] + segment * (
opacity * mask_color[1] + (1 - opacity) * image_data[:, :, 1])
image_mask[:, :, 2] = (1 - segment) * image_data[:, :, 2] + segment * (
opacity * mask_color[2] + (1 - opacity) * image_data[:, :, 2])
plt.clf() # clear image
plt.text(len(image_data[0]) // 2 - 10, -6, CategoryNames[pred_classes_r[0]], fontsize=15)
plt.imshow(image_mask.astype(np.uint8))
print("")
pass
except Exception:
print("..................")
print("...... close .....")
print("..................")
pass
pass
coords = [];
def tellme(s):
print(s)
plt.title(s, fontsize=16)
plt.draw()
#plt.waitforbuttonpress()
if manually == 'no':
fig = plt.figure(1);plt.plot(x,y);plt.yscale("log");plt.ylabel('EQE');plt.xlabel('energy [eV]')#;cid = fig.canvas.mpl_connect('button_press_event', onclick)
while True:
pts = []
while len(pts) < 2:
tellme('Select two points for fit')
pts = np.asarray(plt.ginput(2))
#if len(pts) < 2:
# tellme('Too few points, starting over')
# time.sleep(3) # Wait a second
ph = plt.plot(pts[:, 0], pts[:, 1], '*r', lw=2)
tellme('Happy? press (any) Key for yes,\n mouse click to start over')
coords = pts
if plt.waitforbuttonpress():
plt.close('all')
break
else:
t0 = at[t]
print(t0)
#--------------------------2---------------------------------------------
# map view, select boundaries of seismicity
#------------------------------------------------------------------------
plt.figure(1)
ax1 = plt.subplot(111)
#:TODO plot wells
#???
#:TODO plot seismicity
#???
print("Please click %i times"%( dPar['nClicks']))
tCoord = plt.ginput( dPar['nClicks'])
print("clicked", tCoord)
plt.show()
aLon = np.array( tCoord).T[0]
aLat = np.array( tCoord).T[1]
# project into equal area coordinate system
lon_0, lat_0 = .5*( dPar['xmin']+dPar['xmax']), .5*( dPar['ymin']+dPar['ymax'])
m = Basemap(llcrnrlon = dPar['xmin'], urcrnrlon=dPar['xmax'],
llcrnrlat = dPar['ymin'], urcrnrlat=dPar['ymax'],
projection=dPar['projection'], lon_0 = lon_0, lat_0 = lat_0)
#TODO: project into equal area coordinate system
#???
def on_key( event ):
if event.key=="q" or event.key=="Q":
check.lines[0][0].set_visible(True)
check.lines[0][1].set_visible(True)
check.lines[1][0].set_visible(False)
check.lines[1][1].set_visible(False)
check.lines[2][0].set_visible(False)
check.lines[2][1].set_visible(False)
check.lines[3][0].set_visible(False)
check.lines[3][1].set_visible(False)
check.labels[0].set_text('')
check.labels[1].set_text('')
check.labels[2].set_text('')
check.labels[3].set_text('')
#print "Can haz dblclikz???"
x = plt.ginput(1)
#print x
difference_list = []
new_text = ""
#new_text1 = "Other Nearby Peaks \n"
current_peaks = cat_reader()
for entry in current_peaks:
inten = entry[1]
freq = entry[0]
qnum_up = entry[2]
qnum_low = entry[3]
if math.fabs(float(entry[0])-float(x[0][0]))<100.0 and math.fabs((10**float(entry[1])-float(x[0][1]))/(10**float(entry[1])))<0.2:
difference_list.append((math.fabs(float(entry[0])-float(x[0][0])),freq,qnum_up,qnum_low,inten))
difference_list.sort()
#print difference_list
marker = 0
new_picked_list = []
try:
for z in picked_list:
if difference_list[0][2]==z[1] and difference_list[0][3]==z[2]:
new_text+="\n removed: \n "+str(difference_list[0][2])+" | "+str(difference_list[0][3])+' | '+str(difference_list[0][1])+' | '+str(difference_list[0][4])
marker = 1
check.labels[0].set_text(str(difference_list[0][2])+" | "+str(difference_list[0][3])+' | '+str(difference_list[0][1])+' | '+str(difference_list[0][4]))
check.labels[0].set_fontsize(8)
if len(difference_list)>1:
#new_text1+="\n"+str(difference_list[1][2])+' - '+str(difference_list[1][3])+" "+str(difference_list[1][1])+' '+str(difference_list[1][4])
check.labels[1].set_text(str(difference_list[1][2])+' | '+str(difference_list[1][3])+" | "+str(difference_list[1][1])+' | '+str(difference_list[1][4]))
check.labels[1].set_fontsize(8)
if len(difference_list)>2:
#new_text1+="\n"+str(difference_list[2][2])+' - '+str(difference_list[2][3])+" "+str(difference_list[2][1])+' '+str(difference_list[2][4])
check.labels[2].set_text(str(difference_list[2][2])+' | '+str(difference_list[2][3])+" | "+str(difference_list[2][1])+' | '+str(difference_list[2][4]))
check.labels[2].set_fontsize(8)
if len(difference_list)>3:
check.labels[3].set_text(str(difference_list[3][2])+' | '+str(difference_list[3][3])+" | "+str(difference_list[3][1])+' | '+str(difference_list[3][4]))
check.labels[3].set_fontsize(8)
#new_text1+="\n"+str(difference_list[3][2])+' - '+str(difference_list[3][3])+" "+str(difference_list[3][1])+' '+str(difference_list[3][4])
#if len(difference_list)>1:
# if math.fabs(float(difference_list[0][1])-float(difference_list[1][1]))<0.50:
# picked_list.append((difference_list[1][1],difference_list[1][2],difference_list[1][3],difference_list[1][4]))
# new_text+="\n removed "+str(difference_list[0][2])+"-"+str(difference_list[0][3])+' '+str(difference_list[0][4])
#print "test1"
else:
new_picked_list.append(z)
del picked_list
global picked_list
picked_list = new_picked_list
if marker==0:
check.labels[0].set_text(str(difference_list[0][2])+" | "+str(difference_list[0][3])+' | '+str(difference_list[0][1])+' | '+str(difference_list[0][4]))
check.labels[0].set_fontsize(8)
picked_list.append((difference_list[0][1],difference_list[0][2],difference_list[0][3],difference_list[0][4]))
new_text+="\n added: \n "+str(difference_list[0][2])+" | "+str(difference_list[0][3])+' | '+str(difference_list[0][1])+' | '+str(difference_list[0][4])
if len(difference_list)>1:
#new_text1+="\n"+str(difference_list[1][2])+' - '+str(difference_list[1][3])+" "+str(difference_list[1][1])+' '+str(difference_list[1][4])
check.labels[1].set_text(str(difference_list[1][2])+' | '+str(difference_list[1][3])+" | "+str(difference_list[1][1])+' | '+str(difference_list[1][4]))
check.labels[1].set_fontsize(8)
if len(difference_list)>2:
#new_text1+="\n"+str(difference_list[2][2])+' - '+str(difference_list[2][3])+" "+str(difference_list[2][1])+' '+str(difference_list[2][4])
check.labels[2].set_text(str(difference_list[2][2])+' | '+str(difference_list[2][3])+" | "+str(difference_list[2][1])+' | '+str(difference_list[2][4]))
check.labels[2].set_fontsize(8)
if len(difference_list)>3:
check.labels[3].set_text(str(difference_list[3][2])+' | '+str(difference_list[3][3])+" | "+str(difference_list[3][1])+' | '+str(difference_list[3][4]))
check.labels[3].set_fontsize(8)
#new_text1+="\n"+str(difference_list[3][2])+' - '+str(difference_list[3][3])+" "+str(difference_list[3][1])+' '+str(difference_list[3][4])
#if float(difference_list[0][1])-float(difference_list[1][1])<0.010 and difference_list[0][4]==difference_list[1][4]:
# picked_list.append((difference_list[1][1],difference_list[1][2],difference_list[1][3],difference_list[1][4]))
# new_text+="\n added "+str(difference_list[1][2])+"-"+str(difference_list[1][3])+' '+str(difference_list[1][4])
data = cat_reader()
s_b = []
t_b = []
for x in data:
#if x[2]==difference_list[0][2] and x[3]==difference_list[0][3] and marker==0:
# s_b.append(0.0)
# s_b.append(float(10**float(difference_list[0][4])))
# s_b.append(0.0)
# t_b.append(float(difference_list[0][1])-0.0001)
# t_b.append(difference_list[0][1])
# t_b.append(float(difference_list[0][1])+0.0001)
for y in picked_list:
if x[2]==y[1] and x[3]==y[2]:
s_b.append(0.0)
s_b.append(str(10**float(x[1])))
s_b.append(0.0)
t_b.append(float(x[0])-0.0001)
t_b.append(x[0])
t_b.append(float(x[0])+0.0001)
picked_plt.set_data(t_b,s_b)
text_box.set_text(new_text)
text_box.set_fontsize(10)
#text_box2.set_text(new_text1)
#picked_plt.set_xlim([f_lower_g,f_upper_g])
#print new_text
plt.draw()
#print s_b,t_b
except IndexError:
pass
def drawPts(self, OBJECTIDs, plot_title=True):
# check to see if pts are loaded
try:
type(self.pts)
except AttributeError:
raise RuntimeError('ERROR: pts not loaded yet')
# if self.log:
# self.logger.info('drawPts start...')
user_vi = self.cfg.user_vi
doy_limits = self.cfg.param_nita['doy_limits']
value_limits = self.cfg.param_nita['value_limits']
date_limits = self.cfg.param_nita['date_limits']
if OBJECTIDs == [9999]:
OBJECTIDs = list(set(self.pts['OBJECTID']))
OBJECTIDs = OBJECTIDs[0:25]
handdraw_trajs = []
for OBJECTID in OBJECTIDs:
fig, ax = plt.subplots()
plot_y = self.pts.loc[self.pts['OBJECTID'] ==
OBJECTID][user_vi].values
plot_x = self.pts.loc[self.pts['OBJECTID'] ==
OBJECTID]['date_dist'].values
plot_doy = self.pts.loc[self.pts['OBJECTID'] ==
OBJECTID]['doy'].values
if plot_title:
info_line = self.ref_pts.loc[self.ref_pts['OBJECTID'] ==
OBJECTID]
title = ''.join([
str(item) + ' '
for item in list(info_line.values.flatten())
])
else:
title = ''
plot_x, plot_y, plot_doy = nf.filterLimits(plot_x, plot_y,
plot_doy, value_limits,
date_limits, doy_limits)
doy_limit = doy_limits[0]
mappable = ax.scatter(plot_x,
plot_y,
c=plot_doy,
vmin=doy_limit[0],
vmax=doy_limit[1])
ax.set_xlim([plot_x.min(), plot_x.max()])
ax.set_ylim([plot_y.min(), plot_y.max()])
ax.set_title(title)
plt.ylabel('VI Units')
#
xticks_lables = ax.get_xticks().tolist()
xticks_lables = [
str(xticks_label)[0:4] for xticks_label in xticks_lables
]
ax.set_xticklabels(xticks_lables)
#
fig.colorbar(mappable, label='Day of year')
ginput_res = plt.ginput(n=-1, timeout=0)
handdraw_traj = {'OBJECTID': OBJECTID, 'traj': ginput_res}
handdraw_trajs.append(handdraw_traj)
self.handdraw_trajs = handdraw_trajs
plt.close('all')
if self.log:
self.logger.info('total {} OBJECTIDs drew: ' +
str(OBJECTIDs).format(len(OBJECTIDs)))
self.logger.info('drawPts end...')
from PIL import Image as Img
from pylab import ginput, plot
import homography
import numpy as ny
import scipy
import matplotlib.pyplot as pplot
faceImg = ny.asarray(Img.open('IMG_0519.png').convert('L'))
billImg = ny.asarray(Img.open('billboard.png').convert('L'))
pplot.figure(1)
pplot.imshow(faceImg)
facePts = ny.asarray(pplot.ginput(4)).T
pplot.figure(2)
pplot.imshow(billImg)
billPts = ny.asarray(pplot.ginput(4)).T
pplot.show()
H = homography.H_from_from_point(facePts, billPts)
print('Homography Matrix')
print(H)
maxCount = countMost[i]
mostEle = i
return mostEle
#baseline=mostFreNum(ys)
# newYSS=[]
# newS=[]
# for i in ys:
# newYSS.append(round(i, 6))
# print(mostFreNum(ys))
# print(mostFreNum(newYSS))
# baseline=mostFreNum(newYSS)
#print(baseline)
point = plt.ginput(1)
yaxis = point[0][1]
baseline = yaxis
print(yaxis)
#baseline=float(input("Base on the graph, please enter a y value to set up your baseline: "))
amplitudeBase = float(
input(
"please input minimum peak amplitude(the height from your input baseline): "
))
for i in range(counter):
yzero.append(baseline)
for i in range(counter):
yzero4Peak.append(xs[i] * 0 + baseline)
for i in range(counter):
yzero4Deep.append(xs[i] * 0 + baseline)
plt.plot(xs, yzero, 'r--')
def calculate(data, percent):
#计算距离
distlist = []
dist = np.zeros((len(data), len(data)))
for i in range(len(data) - 1):
for j in range(i + 1, len(data)):
distance = np.sqrt(np.sum(np.square(data[i] - data[j])))
dist[i, j] = distance
dist[j, i] = distance
if (i != j):
distlist.append(distance)
# dc
# for i in range(len(dist) - 1):
# for j in range(len(dist) - 1):
# if(dist[i,j] != 0):
sortdist = sorted(distlist)
position = round(len(distlist) * percent / 100)
dc = sortdist[position]
print('dc', dc)
#计算局部密度
# 计算局部密度 rho (利用 Gaussian 核)
rho = np.zeros(len(dist))
# Gaussian kernel
for i in range(len(dist) - 1):
for j in range(i + 1, len(dist)):
rho[i] = rho[i] + np.exp(-(dist[i, j] / dc) * (dist[i, j] / dc))
rho[j] = rho[j] + np.exp(-(dist[i, j] / dc) * (dist[i, j] / dc))
rho = [item / max(rho) for item in rho]
# 生成 delta 和 nneigh 数组
# delta 距离数组
# nneigh 比本身密度大的最近的点
# 记录 rho 值更大的数据点中与 ordrho(ii) 距离最近的点的编号 ordrho(jj)
# 将 rho 按降序排列，ordrho 保持序
ordrho = np.flipud(np.argsort(rho))
delta = np.zeros(len(dist))
nneigh = np.zeros(len(dist), dtype=int)
delta[ordrho[0]] = -1.
nneigh[ordrho[0]] = 0
# 求最大距离
maxd = max(dist.flatten())
for ii in range(1, len(dist)):
delta[ordrho[ii]] = maxd
for jj in range(ii):
if dist[ordrho[ii], ordrho[jj]] < delta[ordrho[ii]]:
delta[ordrho[ii]] = dist[ordrho[ii], ordrho[jj]]
nneigh[ordrho[ii]] = ordrho[jj]
delta[ordrho[0]] = max(delta)
# delta = [item/max(delta) for item in delta]
gamma = [rho[i] * delta[i] for i in range(len(dist))]
plt.figure(1)
for index in range(len(dist)):
plt.plot(rho[index], delta[index], 'ko')
pos = plt.ginput(1)
rhomin = pos[0][0]
deltamin = pos[0][1]
cl = -1 * np.ones(len(dist), dtype=int)
icl = []
NCLUST = 1
for i in range(len(dist)):
if rho[i] > rhomin and delta[i] > deltamin:
cl[i] = NCLUST #第 i 号数据点属于第 NCLUST 个 cluster
icl.append(i) # 逆映射,第 NCLUST 个 cluster 的中心为第 i 号数据点
NCLUST += 1
for i in range(len(dist)):
if cl[ordrho[i]] == -1:
cl[ordrho[i]] = cl[nneigh[ordrho[i]]]
print(cl)
color = ['bo', 'co', 'go', 'ko', 'mo', 'ro', 'wo', 'yo']
s = random.sample(color, NCLUST - 1)
plt.close()
for i in range(len(data)):
plt.plot(data[i][0], data[i][1], s[cl[i] - 1])
plt.show()
plt.clf()
plt.axis([-1., 1., -1., 1.])
plt.setp(plt.gca(), autoscale_on=False)
tellme('You will define a triangle, click to begin')
plt.waitforbuttonpress()
happy = False
while not happy:
pts = []
while len(pts) < 3:
tellme('Select 3 corners with mouse')
pts = np.asarray(plt.ginput(3, timeout=-1))
if len(pts) < 3:
tellme('Too few points, starting over')
time.sleep(1) # Wait a second
ph = plt.fill(pts[:, 0], pts[:, 1], 'r', lw=2)
tellme('Happy? Key click for yes, mouse click for no')
happy = plt.waitforbuttonpress()
# Get rid of fill
if not happy:
for p in ph:
p.remove()
3)[:, :, :2]
imNp = imrgbn
data_marca = imNp[np.where(np.all(np.equal(markImg, (255, 0, 0)), 2))]
data_fondo = imNp[np.where(np.all(np.equal(markImg, (0, 255, 0)), 2))]
data_linea = imNp[np.where(np.all(np.equal(markImg, (0, 0, 255)), 2))]
plt.figure()
plt.plot(data_marca[:, 0], data_marca[:, 1], 'r.', label='marca')
plt.plot(data_fondo[:, 0], data_fondo[:, 1], 'g.', label='fondo')
plt.plot(data_linea[:, 0], data_linea[:, 1], 'b.', label='linea')
plt.title('Espacio RGB normalizado')
plt.show()
# Primera linea
p12 = plt.ginput(2, timeout=-1, show_clicks=True, mouse_pop=2, mouse_stop=3)
p1 = np.array(p12[0] + (1, ))
p2 = np.array(p12[1] + (1, ))
p = np.array(p12)
l1 = np.cross(p1, p2)
plt.plot(p[:, 0], p[:, 1], 'r-')
# segunda linea
p12 = plt.ginput(2, timeout=-1, show_clicks=True, mouse_pop=2, mouse_stop=3)
p1 = np.array(p12[0] + (1, ))
p2 = np.array(p12[1] + (1, ))
l2 = np.cross(p1, p2)
p = np.array(p12)
plt.plot(p[:, 0], p[:, 1], 'g-')
parser = argparse.ArgumentParser(description='Image for process.')
parser.add_argument('path',
metavar='im',
type=str,
help='Path to image to use in the adjust process')
args = parser.parse_args()
im1 = np.array(Image.open(os.path.abspath(args.path)).\
convert('L'))
plt.figure('Imagen 1')
plt.axis('off'), plt.imshow(im1, cmap='gray')
x1 = np.array(plt.ginput(4, mouse_add=3, mouse_pop=1, mouse_stop=2)).T
plt.close()
h, w = 400, 300
x2 = np.array([[0, w, 0, w], [0, 0, h, h]])
xh1 = make_homog(x1)
xh2 = make_homog(x2)
H = H_from_points(xh2, xh1)
im_out = Htransform(im1, H, (h, w))
plt.figure()
plt.subplot(121), plt.imshow(im1, cmap='gray'), plt.axis('off')
plt.plot(x1[0], x1[1], 'r.', markersize=6)
plt.subplot(122), plt.imshow(im_out, cmap='gray'), plt.axis('off')
plt.show()
def womcat(hop):
"""concatenate data with overlapping wavelength regions"""
import numpy as np
import logging
import matplotlib.pyplot as plt
from tmath.wombat.inputter import inputter
from tmath.wombat.inputter_single import inputter_single
from tmath.wombat.womget_element import womget_element
from tmath.wombat.yesno import yesno
from tmath.wombat.get_screen_size import get_screen_size
from tmath.wombat.womwaverange2 import womwaverange2
from tmath.wombat import HOPSIZE
from matplotlib.widgets import Cursor
plt.ion()
screen_width, screen_height = get_screen_size()
screenpos = '+{}+{}'.format(int(screen_width * 0.2),
int(screen_height * 0.05))
fig = plt.figure()
ax = fig.add_subplot(111)
cursor = Cursor(ax, useblit=True, color='k', linewidth=1)
fig.canvas.manager.window.wm_geometry(screenpos)
fig.canvas.set_window_title('Cat')
fig.set_size_inches(9, 6)
# turns off key stroke interaction
fig.canvas.mpl_disconnect(fig.canvas.manager.key_press_handler_id)
print("\nThis will combine blue and red pieces from two hoppers\n")
hopnum1 = 0
hopnum2 = 0
while (hopnum1 < 1) or (hopnum1 > HOPSIZE):
hopnum1 = inputter('Enter first hopper: ', 'int', False)
while (hopnum2 < 1) or (hopnum2 > HOPSIZE):
hopnum2 = inputter('Enter second hopper: ', 'int', False)
if (hop[hopnum1].wave[0] > hop[hopnum2].wave[0]):
hopnum1, hopnum2 = hopnum2, hopnum1
wdelt1 = hop[hopnum1].wave[1] - hop[hopnum1].wave[0]
wdelt2 = hop[hopnum2].wave[1] - hop[hopnum2].wave[0]
# check if wavelength dispersion same
if (abs(wdelt1 - wdelt2) > 0.00001):
print('Spectra do not have same Angstrom/pixel')
print('Blue side: {}'.format(wdelt1))
print('Red side: {}'.format(wdelt2))
return hop
if hop[hopnum1].wave[-1] < hop[hopnum2].wave[0]:
print('Spectra do not overlap\n')
return hop
print("\nOverlap range is {} to {}".format(hop[hopnum2].wave[0],
hop[hopnum1].wave[-1]))
print("\nPlotting blue side as blue, red side as red\n")
waveblue = hop[hopnum1].wave.copy()
fluxblue = hop[hopnum1].flux.copy()
varblue = hop[hopnum1].var.copy()
wavered = hop[hopnum2].wave.copy()
fluxred = hop[hopnum2].flux.copy()
varred = hop[hopnum2].var.copy()
indexblue = womget_element(waveblue, wavered[0])
indexred = womget_element(wavered, waveblue[-1])
fluxcor = 1.0
blue_mean = np.mean(fluxblue[indexblue:])
red_mean = np.mean(fluxred[0:indexred + 1])
if (blue_mean / red_mean < 0.8) or (blue_mean / red_mean > 1.2):
fluxcor = blue_mean / red_mean
print("Averages very different, scaling red to blue for plot")
print("Red multiplied by {}".format(fluxcor))
plt.cla()
plt.plot(waveblue[indexblue:],
fluxblue[indexblue:],
drawstyle='steps-mid',
color='b')
plt.plot(wavered[0:indexred],
fluxred[0:indexred] * fluxcor,
drawstyle='steps-mid',
color='r')
plt.xlabel('Wavelength')
plt.ylabel('Flux')
plt.pause(0.01)
print('Change scale?')
answer = yesno('n')
if (answer == 'y'):
xmin_old, xmax_old = plt.xlim()
ymin_old, ymax_old = plt.ylim()
done = False
while (not done):
plt.xlim([xmin_old, xmax_old])
plt.ylim([ymin_old, ymax_old])
print('Click corners of box to change plot scale')
newlims = plt.ginput(2, timeout=-1)
xmin = newlims[0][0]
ymin = newlims[0][1]
xmax = newlims[1][0]
ymax = newlims[1][1]
plt.xlim([xmin, xmax])
plt.ylim([ymin, ymax])
print('Is this OK?')
loopanswer = yesno('y')
if (loopanswer == 'y'):
done = True
print('\nEnter method to select wavelength ranges\n')
mode = inputter_single(
'Enter (w)avelengths or mark with the (m)ouse? (w/m) ', 'wm')
print('\nChoose end points of region to compute average\n')
waveb, waver, mode = womwaverange2(waveblue[indexblue:],
fluxblue[indexblue:],
wavered[0:indexred],
fluxred[0:indexred] * fluxcor, mode)
indexblueb = womget_element(waveblue, waveb)
indexbluer = womget_element(waveblue, waver)
indexredb = womget_element(wavered, waveb)
indexredr = womget_element(wavered, waver)
mean_blue = np.mean(fluxblue[indexblueb:indexbluer + 1])
mean_red = np.mean(fluxred[indexredb:indexredr + 1])
print("\nAverage for {}:{}".format(waveb, waver))
print("Blue side: {}".format(mean_blue))
print("Red side: {}\n".format(mean_red))
brscale = inputter_single('Scale to blue or red (b/r)? ', 'br')
if (brscale == 'b'):
brscalefac = mean_blue / mean_red
logging.info('Cat scaling to blue by {}'.format(brscalefac))
fluxred = fluxred * brscalefac
varred = varred * brscalefac**2
else:
brscalefac = mean_red / mean_blue
logging.info('Cat scaling to red by {}'.format(brscalefac))
fluxblue = fluxblue * brscalefac
varblue = varblue * brscalefac**2
# print("\nPlotting blue side as blue, red side as red\n")
# plt.cla()
# plt.plot(waveblue[indexblueb:indexbluer+1],fluxblue[indexblueb:indexbluer+1],drawstyle='steps-mid',color='b')
# plt.plot(wavered[indexredb:indexredr+1],fluxred[indexredb:indexredr+1],drawstyle='steps-mid',color='r')
# plt.xlabel('Wavelength')
# plt.ylabel('Flux')
# plt.pause(0.01)
# print('Change scale?')
# answer=yesno('n')
# if (answer == 'y'):
# xmin_old,xmax_old=plt.xlim()
# ymin_old,ymax_old=plt.ylim()
# done=False
# while (not done):
# plt.xlim([xmin_old,xmax_old])
# plt.ylim([ymin_old,ymax_old])
# print('Click corners of box to change plot scale')
# newlims=plt.ginput(2,timeout=-1)
# xmin=newlims[0][0]
# ymin=newlims[0][1]
# xmax=newlims[1][0]
# ymax=newlims[1][1]
# plt.xlim([xmin,xmax])
# plt.ylim([ymin,ymax])
# print('Is this OK?')
# loopanswer=yesno('y')
# if (loopanswer == 'y'):
# done=True
# print('\nChoose end points of region to compute average\n')
# waveb,waver,mode=womwaverange2(waveblue[indexblueb:indexbluer+1],
# fluxblue[indexblueb:indexbluer+1],
# wavered[indexredb:indexredr+1],
# fluxred[indexredb:indexredr+1],mode)
# indexblueb=womget_element(waveblue,waveb)
# indexbluer=womget_element(waveblue,waver)
# indexredb=womget_element(wavered,waveb)
# indexredr=womget_element(wavered,waver)
# ewadd=inputter_single('Add overlap region (e)qually or with (w)eights (e/w)?','ew')
# if (ewadd == 'e'):
# overflux=(fluxblue[indexblueb:indexbluer+1]+fluxred[indexredb:indexredr+1])/2.
# overvar=(varblue[indexblueb:indexbluer+1]+varred[indexredb:indexredr+1])
#replacing with inverse variance weighted average
delta = (indexbluer - indexblueb) - (indexredr - indexredb)
if delta != 0:
indexredr = indexredr + delta
overflux = np.average([
fluxblue[indexblueb:indexbluer + 1], fluxred[indexredb:indexredr + 1]
],
weights=[
1. / varblue[indexblueb:indexbluer + 1],
1. / varred[indexredb:indexredr + 1]
],
axis=0)
overvar = np.sum(
[varblue[indexblueb:indexbluer + 1], varred[indexredb:indexredr + 1]],
axis=0)
logging.info('Cat combined sides with inverse variance weighted average')
# else:
# wei_done = False
# while (not wei_done):
# weiblue=inputter('Enter fractional weight for blue side: ','float',False)
# weired=inputter('Enter fractional weight for red side: ','float',False)
# if (np.abs((weiblue+weired)-1.0) > 0.000001):
# print('Weights do not add to 1.0')
# else:
# wei_done = True
# overflux=(fluxblue[indexblueb:indexbluer+1]*weiblue+
# fluxred[indexredb:indexredr+1]*weired)
# overvar=(varblue[indexblueb:indexbluer+1]*weiblue**2+
# varred[indexredb:indexredr+1]*weired**2)
# logging.info('Cat adds blue side with weight {} and red side with weight {}'.format(weiblue, weired))
newbluewave = waveblue[:indexblueb]
newblueflux = fluxblue[:indexblueb]
newbluevar = varblue[:indexblueb]
overwave = waveblue[indexblueb:indexbluer + 1]
newredwave = wavered[indexredr + 1:]
newredflux = fluxred[indexredr + 1:]
newredvar = varred[indexredr + 1:]
newwave = np.concatenate([newbluewave, overwave, newredwave])
newflux = np.concatenate([newblueflux, overflux, newredflux])
newvar = np.concatenate([newbluevar, overvar, newredvar])
logging.info('File {} and'.format(hop[hopnum1].obname))
logging.info('File {} concatenated'.format(hop[hopnum2].obname))
logging.info('over wavelength range {} to {}'.format(
waveblue[indexblueb], waveblue[indexbluer]))
plt.clf()
axarr = fig.subplots(2)
axarr[0].plot(overwave, overflux, drawstyle='steps-mid', color='k')
axarr[0].plot(waveblue[indexblueb:indexbluer + 1],
fluxblue[indexblueb:indexbluer + 1],
drawstyle='steps-mid',
color='b')
axarr[0].plot(wavered[indexredb:indexredr + 1],
fluxred[indexredb:indexredr + 1],
drawstyle='steps-mid',
color='r')
axarr[0].set_title('Overlap region, with inputs and combination')
axarr[1].set_ylabel('Flux')
axarr[1].plot(newwave, newflux, drawstyle='steps-mid', color='k')
axarr[1].plot(newwave[indexblueb:indexbluer + 1],
newflux[indexblueb:indexbluer + 1],
drawstyle='steps-mid',
color='r')
plt.pause(0.01)
hopout = 0
while (hopout < 1) or (hopout > HOPSIZE):
hopout = inputter('Enter hopper to store combined spectrum: ', 'int',
False)
hop[hopout].wave = newwave
hop[hopout].flux = newflux
hop[hopout].var = newvar
hop[hopout].obname = hop[hopnum1].obname
hop[hopout].header = hop[hopnum1].header
plt.close()
#fig.clf()
#plt.cla()
return hop
# -*- coding: utf-8 -*-
# CLAB3
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
import time
import cv2
I = Image.open('stereo2012a.jpg')
plt.imshow(I)
uv = plt.ginput(12) # Graphical user interface to get 6 points
#xyz coordinates (world coordinates)
xyz = np.asarray([[7, 7, 0], [14, 7, 0], [21, 7, 0], [21, 14, 0], [0, 7, 7],
[0, 14, 7], [0, 21, 7], [0, 21, 14], [7, 0, 7], [7, 0, 14],
[7, 0, 21], [14, 0, 21]])
'''
%% TASK 1: CALIBRATE
%
% Function to perform camera calibration
%
% Usage: calibrate(image, XYZ, uv)
% return C
% Where: image - is the image of the calibration target.
% XYZ - is a N x 3 array of XYZ coordinates
% of the calibration target points.
% uv - is a N x 2 array of the image coordinates
% of the calibration target points.
% K - is the 3 x 4 camera calibration matrix.
% The variable N should be an integer greater than or equal to 6.
%
def rmshadows(humfile, sonpath, win=31, shadowmask=0, doplot=1):
'''
Remove dark shadows in scans caused by shallows, shorelines, and attenuation of acoustics with distance
Manual or automated processing options available
Works on the radiometrically corrected outputs of the correct module
Syntax
----------
[] = PyHum.rmshadows(humfile, sonpath, win, shadowmask, doplot)
Parameters
----------
humfile : str
path to the .DAT file
sonpath : str
path where the *.SON files are
win : int, *optional* [Default=100]
window size (pixels) for the automated shadow removal algorithm
shadowmask : int, *optional* [Default=0]
1 = do manual shadow masking, otherwise do automatic shadow masking
doplot : int, *optional* [Default=1]
1 = make plots, otherwise do not
Returns
-------
sonpath+base+'_data_star_la.dat': memory-mapped file
contains the starboard scan with water column removed and
radiometrically corrected, and shadows removed
sonpath+base+'_data_port_la.dat': memory-mapped file
contains the portside scan with water column removed and
radiometrically corrected, and shadows removed
'''
# prompt user to supply file if no input file given
if not humfile:
print 'An input file is required!!!!!!'
Tk().withdraw(
) # we don't want a full GUI, so keep the root window from appearing
humfile = askopenfilename(filetypes=[("DAT files", "*.DAT")])
# prompt user to supply directory if no input sonpath is given
if not sonpath:
print 'A *.SON directory is required!!!!!!'
Tk().withdraw(
) # we don't want a full GUI, so keep the root window from appearing
sonpath = askdirectory()
# print given arguments to screen and convert data type where necessary
if humfile:
print 'Input file is %s' % (humfile)
if sonpath:
print 'Sonar file path is %s' % (sonpath)
if win:
win = np.asarray(win, int)
print 'Window is %s square pixels' % (str(win))
if shadowmask:
shadowmask = np.asarray(shadowmask, int)
if shadowmask == 1:
print 'Shadow masking is manual'
else:
print 'Shadow masking is auto'
if doplot:
doplot = int(doplot)
if doplot == 0:
print "Plots will not be made"
# start timer
if os.name == 'posix': # true if linux/mac or cygwin on windows
start = time.time()
else: # windows
start = time.clock()
# if son path name supplied has no separator at end, put one on
if sonpath[-1] != os.sep:
sonpath = sonpath + os.sep
base = humfile.split('.DAT') # get base of file name for output
base = base[0].split(os.sep)[-1]
base = humutils.strip_base(base)
meta = loadmat(os.path.normpath(os.path.join(sonpath, base + 'meta.mat')))
# load memory mapped scans
shape_port = np.squeeze(meta['shape_port'])
if shape_port != '':
#port_fp = np.memmap(sonpath+base+'_data_port_la.dat', dtype='float32', mode='r', shape=tuple(shape_port))
with open(
os.path.normpath(
os.path.join(sonpath, base + '_data_port_la.dat')),
'r') as ff:
port_fp = np.memmap(ff,
dtype='float32',
mode='r',
shape=tuple(shape_port))
shape_star = np.squeeze(meta['shape_star'])
if shape_star != '':
#star_fp = np.memmap(sonpath+base+'_data_star_la.dat', dtype='float32', mode='r', shape=tuple(shape_star))
with open(
os.path.normpath(
os.path.join(sonpath, base + '_data_star_la.dat')),
'r') as ff:
star_fp = np.memmap(ff,
dtype='float32',
mode='r',
shape=tuple(shape_star))
dist_m = np.squeeze(meta['dist_m'])
ft = 1 / (meta['pix_m'])
extent = shape_star[1]
if shadowmask == 1: #manual
Zt = []
if len(np.shape(star_fp)) > 2:
for p in xrange(len(star_fp)):
raw_input(
"Shore picking (starboard), are you ready? 30 seconds. Press Enter to continue..."
)
shoreline_star = {}
fig = plt.figure()
ax = plt.gca()
ax.imshow(star_fp[p], cmap='gray') #, origin = 'upper') #im =
plt.axis('normal')
plt.axis('tight')
pts1 = plt.ginput(
n=300,
timeout=30) # it will wait for 200 clicks or 30 seconds
x1 = map(lambda x: x[0],
pts1) # map applies the function passed as
y1 = map(lambda x: x[1],
pts1) # first parameter to each element of pts
shoreline_star = np.interp(np.r_[:np.shape(star_fp[p])[1]], x1,
y1)
plt.close()
del fig
star_mg = star_fp[p].copy()
shoreline_star = np.asarray(shoreline_star, 'int')
# shift proportionally depending on where the bed is
for k in xrange(np.shape(star_mg)[1]):
star_mg[shoreline_star[k]:, k] = np.nan
del shoreline_star
Zt.append(star_mg)
else:
raw_input(
"Shore picking (starboard), are you ready? 30 seconds. Press Enter to continue..."
)
shoreline_star = {}
fig = plt.figure()
ax = plt.gca()
ax.imshow(star_fp, cmap='gray') #, origin = 'upper') #im =
plt.axis('normal')
plt.axis('tight')
pts1 = plt.ginput(
n=300, timeout=30) # it will wait for 200 clicks or 30 seconds
x1 = map(lambda x: x[0],
pts1) # map applies the function passed as
y1 = map(lambda x: x[1],
pts1) # first parameter to each element of pts
shoreline_star = np.interp(np.r_[:np.shape(star_fp)[1]], x1, y1)
plt.close()
del fig
star_mg = star_fp.copy()
shoreline_star = np.asarray(shoreline_star, 'int')
# shift proportionally depending on where the bed is
for k in xrange(np.shape(star_mg)[1]):
star_mg[shoreline_star[k]:, k] = np.nan
del shoreline_star
Zt.append(star_mg)
## create memory mapped file for Z
#p = np.memmap(sonpath+base+'_data_star_la.dat', dtype='float32', mode='w+', shape=np.shape(Zt))
#fp[:] = Zt[:]
#del fp
Zt = np.squeeze(Zt)
# create memory mapped file for Zs
#fp = np.memmap(sonpath+base+'_data_star_lar.dat', dtype='float32', mode='w+', shape=np.shape(Zs))
with open(
os.path.normpath(
os.path.join(sonpath, base + '_data_star_lar.dat')),
'w+') as ff:
fp = np.memmap(ff, dtype='float32', mode='w+', shape=np.shape(Zt))
fp[:] = Zt[:]
del fp
del Zt
#shutil.move(os.path.normpath(os.path.join(sonpath,base+'_data_star_lar.dat')), os.path.normpath(os.path.join(sonpath,base+'_data_star_la.dat')))
Zt = []
if len(np.shape(star_fp)) > 2:
for p in xrange(len(port_fp)):
raw_input(
"Shore picking (port), are you ready? 30 seconds. Press Enter to continue..."
)
shoreline_port = {}
fig = plt.figure()
ax = plt.gca()
ax.imshow(port_fp[p], cmap='gray') #, origin = 'upper') #im =
plt.axis('normal')
plt.axis('tight')
pts1 = plt.ginput(
n=300,
timeout=30) # it will wait for 200 clicks or 30 seconds
x1 = map(lambda x: x[0],
pts1) # map applies the function passed as
y1 = map(lambda x: x[1],
pts1) # first parameter to each element of pts
shoreline_port = np.interp(np.r_[:np.shape(port_fp[p])[1]], x1,
y1)
plt.close()
del fig
port_mg = port_fp[p].copy()
shoreline_port = np.asarray(shoreline_port, 'int')
# shift proportionally depending on where the bed is
for k in xrange(np.shape(port_mg)[1]):
port_mg[shoreline_port[k]:, k] = np.nan
del shoreline_port
Zt.append(port_mg)
else:
raw_input(
"Shore picking (port), are you ready? 30 seconds. Press Enter to continue..."
)
shoreline_port = {}
fig = plt.figure()
ax = plt.gca()
ax.imshow(port_fp, cmap='gray') #, origin = 'upper') #im =
plt.axis('normal')
plt.axis('tight')
pts1 = plt.ginput(
n=300, timeout=30) # it will wait for 200 clicks or 30 seconds
x1 = map(lambda x: x[0],
pts1) # map applies the function passed as
y1 = map(lambda x: x[1],
pts1) # first parameter to each element of pts
shoreline_port = np.interp(np.r_[:np.shape(port_fp)[1]], x1, y1)
plt.close()
del fig
port_mg = port_fp.copy()
shoreline_port = np.asarray(shoreline_port, 'int')
# shift proportionally depending on where the bed is
for k in xrange(np.shape(port_mg)[1]):
port_mg[shoreline_port[k]:, k] = np.nan
del shoreline_port
Zt.append(port_mg)
Zt = np.squeeze(Zt)
## create memory mapped file for Z
#fp = np.memmap(sonpath+base+'_data_port_la.dat', dtype='float32', mode='w+', shape=np.shape(Zt))
#fp[:] = Zt[:]
#del fp
# create memory mapped file for Zp
#fp = np.memmap(sonpath+base+'_data_port_lar.dat', dtype='float32', mode='w+', shape=np.shape(Zp))
with open(
os.path.normpath(
os.path.join(sonpath, base + '_data_port_lar.dat')),
'w+') as ff:
fp = np.memmap(ff, dtype='float32', mode='w+', shape=np.shape(Zt))
fp[:] = Zt[:]
del fp
del Zt
#shutil.move(os.path.normpath(os.path.join(sonpath,base+'_data_port_lar.dat')), os.path.normpath(os.path.join(sonpath,base+'_data_port_la.dat')))
else: #auto
#win = 31
Zs = []
Zp = []
if len(np.shape(star_fp)) > 2:
for p in xrange(len(star_fp)):
merge = np.vstack((np.flipud(port_fp[p]), star_fp[p]))
merge = np.asarray(merge, 'float64')
merge_mask = np.vstack((np.flipud(port_fp[p]), star_fp[p]))
merge[merge_mask == 0] = 0
del merge_mask
mask = np.asarray(merge != 0,
'int8') # only 8bit precision needed
merge[np.isnan(merge)] = 0
#Z,ind = humutils.sliding_window(merge,(win,win),(win/2,win/2))
Z, ind = humutils.sliding_window(merge, (win, win), (win, win))
#zmean = np.reshape(zmean, ( ind[0], ind[1] ) )
Ny, Nx = np.shape(merge)
#zmean[np.isnan(zmean)] = 0
try: #parallel processing with all available cores
w = Parallel(n_jobs=-1,
verbose=0)(delayed(parallel_me)(Z[k])
for k in xrange(len(Z)))
except: #fall back to serial
w = Parallel(n_jobs=1,
verbose=0)(delayed(parallel_me)(Z[k])
for k in xrange(len(Z)))
zmean = np.reshape(w, (ind[0], ind[1]))
del w
M = humutils.im_resize(zmean, Nx, Ny)
M[mask == 0] = 0
del zmean
bw = M > 0.5
del M
# erode and dilate to remove splotches of no data
bw2 = binary_dilation(binary_erosion(bw,
structure=np.ones(
(3, 3))),
structure=np.ones((13, 13)))
#bw2 = binary_dilation(binary_erosion(bw,structure=np.ones((win/4,win/4))), structure=np.ones((win/4,win/4)))
##bw2 = binary_erosion(bw,structure=np.ones((win*2,win*2)))
## fill holes
bw2 = binary_fill_holes(bw2, structure=np.ones(
(win, win))).astype(int)
merge2 = grey_erosion(merge, structure=np.ones((win, win)))
#del bw
#bw2 = np.asarray(bw2!=0,'int8') # we only need 8 bit precision
bw2 = np.asarray(bw != 0,
'int8') # we only need 8 bit precision
del bw
merge[bw2 == 1] = 0 #blank out bad data
merge[merge2 == np.min(merge2)] = 0 #blank out bad data
del merge2
## do plots of merged scans
if doplot == 1:
Zdist = dist_m[shape_port[-1] * p:shape_port[-1] * (p + 1)]
fig = plt.figure()
plt.imshow(merge,
cmap='gray',
extent=[
min(Zdist),
max(Zdist), -extent * (1 / ft),
extent * (1 / ft)
])
plt.ylabel('Range (m)'), plt.xlabel(
'Distance along track (m)')
plt.axis('normal')
plt.axis('tight')
custom_save(sonpath,
'merge_corrected_rmshadow_scan' + str(p))
del fig
Zp.append(np.flipud(merge[:shape_port[1], :]))
Zs.append(merge[shape_port[1]:, :])
del merge, bw2
else:
merge = np.vstack((np.flipud(port_fp), star_fp))
merge = np.asarray(merge, 'float64')
merge_mask = np.vstack((np.flipud(port_fp), star_fp))
merge[merge_mask == 0] = 0
del merge_mask
mask = np.asarray(merge != 0, 'int8') # only 8bit precision needed
merge[np.isnan(merge)] = 0
#Z,ind = humutils.sliding_window(merge,(win,win),(win/2,win/2))
Z, ind = humutils.sliding_window(merge, (win, win), (win, win))
#zmean = np.reshape(zmean, ( ind[0], ind[1] ) )
Ny, Nx = np.shape(merge)
#zmean[np.isnan(zmean)] = 0
try: #parallel processing with all available cores
w = Parallel(n_jobs=-1, verbose=0)(delayed(parallel_me)(Z[k])
for k in xrange(len(Z)))
except: #fall back to serial
w = Parallel(n_jobs=1, verbose=0)(delayed(parallel_me)(Z[k])
for k in xrange(len(Z)))
zmean = np.reshape(w, (ind[0], ind[1]))
del w
M = humutils.im_resize(zmean, Nx, Ny)
M[mask == 0] = 0
del zmean
bw = M > 0.5
del M
# erode and dilate to remove splotches of no data
bw2 = binary_dilation(binary_erosion(bw, structure=np.ones(
(3, 3))),
structure=np.ones((13, 13)))
#bw2 = binary_dilation(binary_erosion(bw,structure=np.ones((win/4,win/4))), structure=np.ones((win/4,win/4)))
##bw2 = binary_erosion(bw,structure=np.ones((win*2,win*2)))
## fill holes
bw2 = binary_fill_holes(bw2, structure=np.ones(
(win, win))).astype(int)
merge2 = grey_erosion(merge, structure=np.ones((win, win)))
#del bw
#bw2 = np.asarray(bw2!=0,'int8') # we only need 8 bit precision
bw2 = np.asarray(bw != 0, 'int8') # we only need 8 bit precision
del bw
merge[bw2 == 1] = 0 #blank out bad data
merge[merge2 == np.min(merge2)] = 0 #blank out bad data
del merge2
# erode and dilate to remove splotches of no data
#bw2 = binary_dilation(binary_erosion(bw,structure=np.ones((3,3))), structure=np.ones((13,13)))
#bw2 = binary_dilation(binary_erosion(bw,structure=np.ones((win,win))), structure=np.ones((win*2,win*2)))
#bw2 = binary_erosion(bw,structure=np.ones((win,win)))
# fill holes
#bw2 = binary_fill_holes(bw2, structure=np.ones((3,3))).astype(int)
#del bw
#bw2 = np.asarray(bw2!=0,'int8') # we only need 8 bit precision
#merge[bw2==1] = 0 #blank out bad data
## do plots of merged scans
if doplot == 1:
Zdist = dist_m
fig = plt.figure()
plt.imshow(merge,
cmap='gray',
extent=[
min(Zdist),
max(Zdist), -extent * (1 / ft),
extent * (1 / ft)
])
plt.ylabel('Range (m)'), plt.xlabel('Distance along track (m)')
plt.axis('normal')
plt.axis('tight')
custom_save(sonpath, 'merge_corrected_rmshadow_scan' + str(0))
del fig
Zp.append(np.flipud(merge[:shape_port[0], :]))
Zs.append(merge[shape_port[0]:, :])
del merge, bw2
Zp = np.squeeze(Zp)
Zs = np.squeeze(Zs)
# create memory mapped file for Zp
#fp = np.memmap(sonpath+base+'_data_port_lar.dat', dtype='float32', mode='w+', shape=np.shape(Zp))
#with open(sonpath+base+'_data_port_lar.dat', 'w+') as f:
with open(
os.path.normpath(
os.path.join(sonpath, base + '_data_port_lar.dat')),
'w+') as ff:
fp = np.memmap(ff, dtype='float32', mode='w+', shape=np.shape(Zp))
fp[:] = Zp[:]
del fp
del Zp
#shutil.move(sonpath+base+'_data_port_lar.dat', sonpath+base+'_data_port_la.dat')
#shutil.move(os.path.normpath(os.path.join(sonpath,base+'_data_port_lar.dat')), os.path.normpath(os.path.join(sonpath,base+'_data_port_la.dat')))
# create memory mapped file for Zs
#fp = np.memmap(sonpath+base+'_data_star_lar.dat', dtype='float32', mode='w+', shape=np.shape(Zs))
#with open(sonpath+base+'_data_star_lar.dat', 'w+') as f:
with open(
os.path.normpath(
os.path.join(sonpath, base + '_data_star_lar.dat')),
'w+') as ff:
fp = np.memmap(ff, dtype='float32', mode='w+', shape=np.shape(Zs))
fp[:] = Zs[:]
del fp
del Zs
#shutil.move(sonpath+base+'_data_star_lar.dat', sonpath+base+'_data_star_la.dat')
#shutil.move(os.path.normpath(os.path.join(sonpath,base+'_data_star_lar.dat')), os.path.normpath(os.path.join(sonpath,base+'_data_star_la.dat')))
if os.name == 'posix': # true if linux/mac
elapsed = (time.time() - start)
else: # windows
elapsed = (time.clock() - start)
print "Processing took ", elapsed, "seconds to analyse"
print "Done!"
def ppicker(eventdir, pname, sname, get, args):
""" Function ppicker is a ginput based arrival picker
allowing for 3 time picks which are saved in SAC data headers"""
pf = os.path.join(eventdir, pname)
sf = os.path.join(eventdir, sname)
pt = read(pf)[0]
st = read(sf)[0]
p, s = (pt.data, st.data)
dt = pt.stats.delta
N = len(pt.data)
b = pt.stats.sac['b']
depth = pt.stats.sac['evdp']
t0 = (pt.stats.sac['t0'] - b) / dt
t4 = (pt.stats.sac['t4'] - b) / dt
t7 = (pt.stats.sac['t7'] - b) / dt
# Skip if T1 and T3 are greater than zero. The default is a large negative number
if args.unpicked:
if float(pt.stats.sac['t1']) > 0 and float(pt.stats.sac['t3']) > 0:
return 's'
left = round(t0 - 15 / dt)
right = round(t0 + 140 / dt)
t = np.around(np.arange(-t0 * dt, (N - t0) * dt, dt)) # Time axis
nn = np.arange(0, N)
plt.figure(num=None, figsize=(22, 6))
plt.plot(p, label='Pcomp')
plt.xticks(nn[::round(5 / dt)], t[::round(5 / dt)]) # Changed from 200
plt.title('{} \n P-trace, source depth = {}'.format(eventdir, depth))
plt.axvline(x=t0, color='y', label='gett P')
plt.axvline(x=t4, color='g', label='gett pP')
if t7 < right:
plt.axvline(x=t7, color='r', label='gett PP')
plt.xlim(left, right)
plt.xlabel('Time \n P arrival is zero seconds')
plt.legend()
print "Keep trace? 'n' for no, 'r' for redo, 'p' for previous, 's' for skip and any other for yes: "
if args.sort:
plt.show(block=False)
inp = get()
plt.close()
return inp
else:
x = plt.ginput(n=2, timeout=0, show_clicks=True)
try:
T1 = x[0][0] * dt + b
T3 = x[1][0] * dt + b
except IndexError:
print "Not all picks made in", eventdir
print "Please retry the picks"
return 'r'
plt.close()
inp = get()
if inp not in ['n', 'r', 'p', 's']:
# Assume yes and save
pt.stats.sac['t1'] = T1
pt.stats.sac['t3'] = T3
pt.write(pf, format='SAC')
st.stats.sac['t1'] = T1
st.stats.sac['t3'] = T3
st.write(sf, format='SAC')
return inp