def motion_detector():
global max_area, avg, prev_pos, largest_contour
contour = cv.FindContours(
temp_image,
store,
mode=cv.CV_RETR_EXTERNAL,
method=cv.CV_CHAIN_APPROX_NONE) #Findling contours
cv.Copy(img, render_image) #Copying for painting on the image
if len(contour) != 0:
temp_contour = contour
area = 0
max_area_test = max_area
while temp_contour != None: #Routine to find the largest contour
area = cv.ContourArea(temp_contour)
if area > max_area_test:
largest_contour = temp_contour
max_area_test = area
temp_contour = temp_contour.h_next()
rect = cv.BoundingRect(largest_contour)
cv.DrawContours(render_image, largest_contour, (0, 255, 0),
(0, 0, 255), 1, 3)
cv.Rectangle(render_image, (rect[0], rect[1]),
(rect[0] + rect[2], rect[1] + rect[3]), (255, 0, 0))
avg = rect[0] + rect[2] / 2
else:
avg = img.width / 2
def _get_pos_countour(self, th_img):
storage = cv.CreateMemStorage(0)
contour = None
ci = cv.CreateImage(cv.GetSize(th_img), 8, 1)
ci = cv.CloneImage(th_img)
contour = cv.FindContours(th_img, storage, cv.CV_RETR_CCOMP, cv.CV_CHAIN_APPROX_SIMPLE)
points = []
while contour:
bound_rect = cv.BoundingRect(list(contour))
contour = contour.h_next()
pt1 = (bound_rect[0], bound_rect[1])
pt2 = (bound_rect[0] + bound_rect[2], bound_rect[1] + bound_rect[3])
points.append(pt1)
points.append(pt2)
#cv.Rectangle(img, pt1, pt2, cv.CV_RGB(255,0,0), 1)
center_point = None
if len(points):
center_point = reduce(lambda a, b: ((a[0] + b[0]) / 2, (a[1] + b[1]) / 2), points)
#cv.Circle(depth, center_point, 10, cv.CV_RGB(255, 255, 255), 1)
return center_point
def findPoints(frame, oldRectPoints):
imageSize = cv.GetSize(frame)
original = cv.CloneImage(frame)
hsv = cv.CreateImage(imageSize, 8, 3)
threshold = cv.CreateImage(imageSize, 8, 1)
# Do things to the image to isolate the red parts
cv.CvtColor(original, hsv, cv.CV_RGB2HSV)
cv.InRangeS(hsv, (110, 80, 80), (140, 255, 255), threshold)
cv.Erode(threshold, threshold, iterations=5)
cv.Dilate(threshold, threshold, iterations=5)
cv.ShowImage("shit", threshold)
memory = cv.CreateMemStorage(0)
clone = cv.CloneImage(threshold)
contours = cv.FindContours(clone, memory, cv.CV_RETR_LIST,
cv.CV_CHAIN_APPROX_SIMPLE, (0, 0))
# area = cv.ContourArea(contours)
if not contours:
# If there's no red on the screen
rectPoints = oldRectPoints
else:
rectPoints = cv.BoundingRect(contours)
# print rectPoints
return rectPoints
def dewarp(image, window, clicked_corners):
debug = cv.CloneImage(image)
# draw red line around edges for debug purposes
cv.PolyLine(debug, [[
clicked_corners[0], clicked_corners[1], clicked_corners[3],
clicked_corners[2]
]], True, cv.RGB(0, 255, 0), 7)
cv.ShowImage(window, debug)
cv.WaitKey()
# Assemble a rotated rectangle out of that info
#rot_box = cv.MinAreaRect2(corners)
enc_box = cv.BoundingRect(clicked_corners)
new_corners = [(0, 0), (enc_box[2] - 1, 0), (0, enc_box[3] - 1),
(enc_box[2] - 1, enc_box[3] - 1)]
warp_mat = cv.CreateMat(3, 3, cv.CV_32FC1)
cv.GetPerspectiveTransform(clicked_corners, new_corners, warp_mat)
rotated = cv.CloneImage(image)
cv.WarpPerspective(image, rotated, warp_mat)
cv.ShowImage(window, rotated)
cv.WaitKey(10)
return rotated
def getStandardizedRects(self):
'''
@return: the boxes centered on the target center of mass +- n_sigma*std
@note: You must call detect() before getStandardizedRects() to see updated results.
'''
#create a list of the top-level contours found in the contours (cv.Seq) structure
rects = []
if len(self._contours) < 1: return (rects)
seq = self._contours
while not (seq == None):
(x, y, w, h) = cv.BoundingRect(seq)
if (cv.ContourArea(seq) >
self._minArea): # and self._filter(rect)
r = pv.Rect(x, y, w, h)
moments = cv.Moments(seq)
m_0_0 = cv.GetSpatialMoment(moments, 0, 0)
m_0_1 = cv.GetSpatialMoment(moments, 0, 1)
m_1_0 = cv.GetSpatialMoment(moments, 1, 0)
mu_2_0 = cv.GetCentralMoment(moments, 2, 0)
mu_0_2 = cv.GetCentralMoment(moments, 0, 2)
cx = m_1_0 / m_0_0
cy = m_0_1 / m_0_0
w = 2.0 * self._rect_sigma * np.sqrt(mu_2_0 / m_0_0)
h = 2.0 * self._rect_sigma * np.sqrt(mu_0_2 / m_0_0)
r = pv.CenteredRect(cx, cy, w, h)
rects.append(r)
seq = seq.h_next()
if self._filter != None:
rects = self._filter(rects)
return rects
def process(self):
def seq_to_iter(seq):
while seq:
yield seq
seq = seq.h_next()
def score_rect(r,p):
x,y,w,h = r
return w * h
cv.Resize(self.frame, self.resized_frame)
cv.CvtColor(self.resized_frame, self.hsv_frame, cv.CV_RGB2HSV)
cv.Smooth(self.hsv_frame, self.smooth_frame, cv.CV_GAUSSIAN,
31)
for p in self.pompon:
if p.calibration_done:
self.in_range(p, self.smooth_frame, self.bin_frame)
cv.Erode(self.bin_frame, self.mask, None, self.dilatation);
if self.show_binary:
self.mask2 = cv.CloneImage(self.mask) # for miniature
contour = seq_to_iter(cv.FindContours(self.mask,
cv.CreateMemStorage(), cv.CV_RETR_EXTERNAL,
cv.CV_CHAIN_APPROX_SIMPLE));
rects = map((lambda c: cv.BoundingRect(c, 0)), contour)
if rects:
x,y,w,h = max(rects, key=lambda r: score_rect(r,p))
p.pos = self.proc2sym(x+w/2,y+h/2)
def findRectPoints(self, oldRectPoints):
hueRange = self.hueRange
clone = cv.CloneImage(self.frame)
hsv = cv.CloneImage(self.channels3)
threshold = cv.CloneImage(self.channels1)
threshold2 = cv.CloneImage(self.channels1)
cv.CvtColor(clone, hsv, cv.CV_RGB2HSV)
cv.InRangeS(hsv, (165, 100, 100), (180, 255, 255), threshold)
cv.InRangeS(hsv, (0, 100, 100), (15, 255, 255), threshold2)
cv.Add(threshold, threshold2, threshold)
self.hue += 1
print self.hue
cv.Erode(threshold, threshold, iterations=5)
cv.Dilate(threshold, threshold, iterations=5)
cv.ShowImage(self.color, threshold)
memory = cv.CreateMemStorage(0)
clone2 = cv.CloneImage(threshold)
contours = cv.FindContours(clone2, memory, cv.CV_RETR_LIST,
cv.CV_CHAIN_APPROX_SIMPLE, (0, 0))
if not contours:
rectPoints = oldRectPoints
else:
rectPoints = cv.BoundingRect(contours)
return rectPoints
def compute_sample_image_coordinates(image_shape, sample_bounds):
# TODO: get this coefficient out of here
# Also, if the sample is not a square, we may amplify the differences by
# the neighborhood, unnaturally. Include a maximum number of pixels as
# a function of screen resolution
neighborhood_size_coeff = Properties.neighborhood_size_coeff
sample_roi = cv.BoundingRect(sample_bounds, 0)
neighborhood_roi = (sample_roi[0] -
neighborhood_size_coeff * sample_roi[2],
sample_roi[1] -
neighborhood_size_coeff * sample_roi[3],
sample_roi[2] * (2 * neighborhood_size_coeff + 1),
sample_roi[3] * (2 * neighborhood_size_coeff + 1))
#print cv.Get1D(backprojected_neighborhood_boundary, 0)
neighborhood_roi = cvutils.rect_intersection(
neighborhood_roi, (0, 0, image_shape[1], image_shape[0]))
replicated_neighborhood_corner = np.hstack((neighborhood_roi[0] * np.ones(
(4, 1)), neighborhood_roi[1] * np.ones((4, 1))))
replicated_neighborhood_corner = replicated_neighborhood_corner.reshape(
(1, 4, 2))
sample_bounds = np.array(sample_bounds)
local_sample_boundary = sample_bounds - replicated_neighborhood_corner
return neighborhood_roi, cvutils.array2point_list(local_sample_boundary)
def find_connected_components(frame):
"""Find connected components from an image.
:: iplimage -> [ dict(box<CvBox2D>, rect<CvRect>) ]
Takes as input a grayscale image that should have some blobs in
it. Outputs a data structure containing the 'centers' of the blobs
and minimal rectangles enclosing them. A maximum of 3 blobs are
returned.
The input frame is modified in place.
"""
contours = get_contours(frame)
#out = draw_contours(frame, contours)
candidates = []
for contour in contours:
storage = cv.CreateMemStorage(0)
minBox = cv.MinAreaRect2(contour, storage)
boundingRect = cv.BoundingRect(contour, 0)
candidates.append({'box': minBox, 'rect': boundingRect})
candidates = sorted(candidates,
key=lambda x: getArea(x['box']),
reverse=True)
return candidates
def findRectPoints(self, oldRectPoints):
hueRange = self.hueRange
satRange = self.satRange
valRange = self.valRange
clone = cv.CloneImage(self.frame)
hsv = cv.CloneImage(self.channels3)
threshold = cv.CloneImage(self.channels1)
threshold2 = cv.CloneImage(self.channels1)
cv.Smooth(clone, clone, cv.CV_GAUSSIAN, 7, 7)
cv.CvtColor(clone, hsv, cv.CV_BGR2HSV)
cv.InRangeS(hsv, (hueRange[0], satRange[0], valRange[0]),
(hueRange[1], satRange[1], satRange[1]), threshold)
cv.InRangeS(hsv, (hueRange[2], satRange[0], satRange[0]),
(hueRange[3], satRange[1], valRange[1]), threshold2)
cv.Add(threshold, threshold2, threshold)
cv.Erode(threshold, threshold, iterations=5)
cv.Dilate(threshold, threshold, iterations=5)
# cv.ShowImage(self.color, threshold)
memory = cv.CreateMemStorage(0)
clone2 = cv.CloneImage(threshold)
contours = cv.FindContours(clone2, memory, cv.CV_RETR_LIST,
cv.CV_CHAIN_APPROX_SIMPLE, (0, 0))
if not contours:
rectPoints = oldRectPoints
else:
rectPoints = cv.BoundingRect(list(contours))
return rectPoints
def find_biggest_region(self):
""" this code should find the biggest region and
then determine some of its characteristics, which
will help direct the drone
"""
# copy the thresholded image
cv.Copy( self.threshed_image, self.copy ) # copy self.threshed_image
# this is OpenCV's call to find all of the contours:
contours = cv.FindContours(self.copy, self.storage, cv.CV_RETR_EXTERNAL,
cv.CV_CHAIN_APPROX_SIMPLE)
# Next we want to find the *largest* contour
if len(contours)>0:
biggest = contours
biggestArea=cv.ContourArea(contours)
while contours != None:
nextArea=cv.ContourArea(contours)
if biggestArea < nextArea:
biggest = contours
biggestArea = nextArea
contours=contours.h_next()
#Use OpenCV to get a bounding rectangle for the largest contour
self.br = cv.BoundingRect(biggest,update=0)
#Publish the data.
self.publishBoxData()
def get_elements(filename, treshold=50, minheight=15, minarea=200, elements=6):
src = cv.LoadImage(filename, cv.CV_LOAD_IMAGE_GRAYSCALE)
test = cv.CreateImage(cv.GetSize(src), 32, 3)
dst = cv.CreateImage(cv.GetSize(src), 8, 1)
storage = cv.CreateMemStorage(0)
cv.Canny(src, dst, treshold, treshold * 3, 3)
storage = cv.CreateMemStorage(0)
seqs = cv.FindContours(dst, storage, cv.CV_RETR_TREE,
cv.CV_CHAIN_APPROX_NONE, (0, 0))
res = []
c = seqs.h_next()
while True:
if not c:
break
box = cv.BoundingRect(c)
area = box[2] * box[3]
#and (area > minarea)
if (box[3] > minheight):
res.append(box)
c = c.h_next()
if len(res) < elements:
while len(res) < elements:
m = 0
c = 0
for i, e in enumerate(res):
if e[3] > m:
m = e[3]
c = i
big = res.pop(c)
res.append((big[0], big[1], int(big[2] * 1.0 / 2), big[3]))
res.append((big[0] + int(big[2] * 1.0 / 2), big[1],
int(big[2] * 1.0 / 2), big[3]))
#for box in res:
# cv.Rectangle(dst, (box[0],box[1]), (box[0]+box[2],box[1]+box[3]), cv.RGB(255,255,255))
#cv.ShowImage('Preview2',dst)
#cv.WaitKey()
imgs = []
print len(res)
for box in res:
cv.SetImageROI(src, box)
tmp = cv.CreateImage((box[2], box[3]), 8, 1)
cv.Copy(src, tmp)
hq.heappush(imgs, (box[0], tmp))
cv.ResetImageROI(src)
res = [hq.heappop(imgs)[1] for i in xrange(len(res))]
return res
def blob_statistics(binary_image,
max_area=99999.0,
max_dim=99999.0): #, show=False):
statistics = []
storage = cv.CreateMemStorage(0)
#FindContours(image, storage, mode=CV_RETR_LIST, method=CV_CHAIN_APPROX_SIMPLE, offset=(0, 0))
contours = cv.FindContours(binary_image, storage, cv.CV_RETR_TREE,
cv.CV_CHAIN_APPROX_SIMPLE, (0, 0))
#number_contours, contours = cv.FindContours(binary_image, storage, cv.sizeof_CvContour, cv.CV_RETR_TREE, cv.CV_CHAIN_APPROX_SIMPLE, (0,0))
#TODO: FIGURE OUT WHAT THE EQUIV OF SIZEOF IS IN OPENCV2
#import pdb
#pdb.set_trace()
original_ptr = contours
while contours != None:
try:
bx, by, bwidth, bheight = cv.BoundingRect(contours, 0)
bounding_rect = Rect(bx, by, bwidth, bheight)
moments = cv.Moments(contours, 0)
#area = moments.m00
#approximation to area since cvMoments' area seem broken
area = bounding_rect.width * bounding_rect.height
if False:
#TODO NOT WORKING!!
if moments.m00 == 0.0:
centroid = (bounding_rect.x, bounding_rect.y)
else:
centroid = (moments.m10 / moments.m00,
moments.m01 / moments.m00)
else:
if bwidth > 0:
cx = bx + bwidth / 2.
else:
cx = bx
if bheight > 0:
cy = by + bheight / 2.
else:
cy = by
centroid = (cx, cy)
#if show:
# print 'areas is', area, bounding_rect.width, bounding_rect.height
if area > max_area or bounding_rect.width > max_dim or bounding_rect.height > max_dim:
cv.DrawContours(binary_image, contours, cv.Scalar(0),
cv.Scalar(0), 0, cv.CV_FILLED)
else:
stats = {
'area': area,
'centroid': centroid,
'rect': bounding_rect
}
statistics.append(stats)
contours = contours.h_next()
except Exception, e:
pass
#This is due to OPENCV BUG and not being able to see inside contour object'
break
def get_blob_from_contour(contour):
bound_rect = cv.BoundingRect(list(contour)) # (x, y, wid, hig)
# get points of longest diagonal
pt1 = (bound_rect[0], bound_rect[1])
pt2 = Math.add_vectors(pt1, (bound_rect[2], bound_rect[3]))
# get centroid by adding them and scaling the result by 1/2
# float is used(*2.0*), otherwise scaling will be by 0
centroid = Math.int_vec(Math.add_vectors(pt1, pt2, 1 / 2.0))
return pt1, pt2, centroid
def seqs_boxes(seqs, minsize=25):
boxes = []
new_seqs = []
for seq in seqs:
box = cv.BoundingRect(seq)
if box[2] >= minsize and box[3] >= minsize and not is_inside(
boxes, box):
boxes.append(box)
new_seqs.append(seq)
boxes = [box for box in boxes if not is_inside(boxes, box)]
return boxes, new_seqs
def find_biggest_region():
""" finds all the contours in threshed image, finds the largest of those,
and then marks in in the main image
"""
# get D so that we can change values in it
global D
cv.Copy(D.threshed_image, D.copy) # copy threshed image
# this is OpenCV's call to find all of the contours:
contours = cv.FindContours(D.copy, D.storage, cv.CV_RETR_EXTERNAL, \
cv.CV_CHAIN_APPROX_SIMPLE)
# Next we want to find the *largest* contour
if len(contours) > 0:
biggest = contours
biggestArea = cv.ContourArea(contours)
while contours != None:
nextArea = cv.ContourArea(contours)
if biggestArea < nextArea:
biggest = contours
biggestArea = nextArea
contours = contours.h_next()
# Use OpenCV to get a bounding rectangle for the largest contour
br = cv.BoundingRect(biggest, update=0)
#print "in find_regions, br is", br
# Example of drawing a red box
# Variables: ulx ~ upper left x, lry ~ lower right y, etc.
ulx = br[0]
lrx = br[0] + br[2]
uly = br[1]
lry = br[1] + br[3]
cv.PolyLine(D.image, [[(ulx, uly), (lrx, uly), (lrx, lry),
(ulx, lry)]], 1, cv.RGB(255, 0, 0))
# Example of drawing a yellow circle
# Variables: cenx, ceny
cenx = (ulx + lrx) / 2
ceny = (uly + lry) / 2
cv.Circle(D.image, (cenx, ceny),
8,
cv.RGB(255, 255, 0),
thickness=1,
lineType=8,
shift=0)
def detecta(imagem):
cv.Smooth(imagem, imagem, cv.CV_GAUSSIAN, 3)
maiorArea = 0
listaContornos = []
listaVertices = []
cv.AbsDiff(imagem, fundo, mascara)
cv.CvtColor(mascara, cinza, cv.CV_BGR2GRAY)
cv.Threshold(cinza, cinza, 50, 255, cv.CV_THRESH_BINARY)
cv.Dilate(cinza, cinza, None, 18)
cv.Erode(cinza, cinza, None, 18)
armazenamento = cv.CreateMemStorage(0)
contorno = cv.FindContours(cinza, armazenamento, cv.CV_RETR_LIST,
cv.CV_LINK_RUNS)
while contorno:
vertices_do_retangulo = cv.BoundingRect(list(contorno))
listaVertices.append(vertices_do_retangulo)
listaContornos.append(cv.ContourArea(contorno))
maiorArea = max(listaContornos)
maiorArea_index = listaContornos.index(maiorArea)
retangulo_de_interesse = listaVertices[maiorArea_index]
contorno = contorno.h_next()
ponto1 = (retangulo_de_interesse[0], retangulo_de_interesse[1])
ponto2 = (retangulo_de_interesse[0] + retangulo_de_interesse[2],
retangulo_de_interesse[1] + retangulo_de_interesse[3])
cv.Rectangle(imagem, ponto1, ponto2, cv.CV_RGB(0, 0, 0), 2)
cv.Rectangle(cinza, ponto1, ponto2, cv.CV_RGB(255, 255, 255), 1)
largura = ponto2[0] - ponto1[0]
altura = ponto2[1] - ponto1[1]
cv.Line(cinza, (ponto1[0] + largura / 2, ponto1[1]),
(ponto1[0] + largura / 2, ponto2[1]), cv.CV_RGB(255, 255,
255), 1)
cv.Line(cinza, (ponto1[0], ponto1[1] + altura / 2),
(ponto2[0], ponto1[1] + altura / 2), cv.CV_RGB(255, 255,
255), 1)
global x
x = ((640 / 2 - (ponto1[0] + (largura / 2))) * -1) / 5
cv.ShowImage("Webcam", imagem)
cv.ShowImage("Mascara", mascara)
cv.ShowImage("Cinza", cinza)
def motion_bbox(output, motion):
#global muestra, prom
global bbox_list
#INICIO = time.time()
contour = cv.FindContours(motion, mem_storage, cv.CV_RETR_CCOMP,
cv.CV_CHAIN_APPROX_SIMPLE)
bbox_list = [] #lista para almacenar los bounding box de las manchas
average_box_area = 0 #variable para obtener el area en promedio de las bounding box
while contour: #recorrido de los contornos/manchas
bbox = cv.BoundingRect(
list(contour)) #obtencion del bounding box del contorno actual
pt1 = (bbox[0], bbox[1]) #punto 1 del bounding box
pt2 = (bbox[0] + bbox[2], bbox[1] + bbox[3]) #punto 2 del bounding box
w, h = abs(pt1[0] - pt2[0]), abs(
pt1[1] - pt2[1]) #ancho y largo del bounding box
#obtencion de puntos del contorno para crear un wire-frame
polygon_points = cv.ApproxPoly(list(contour), mem_storage,
cv.CV_POLY_APPROX_DP)
#mostrar o las manchas de movimiento
if SHOW_MOVEMENT_AREA:
cv.FillPoly(output, [
list(polygon_points),
], cv.CV_RGB(255, 255, 255), 0, 0)
#mostrar o no los contornos de las manchas de movimiento
if SHOW_MOVEMENT_CONTOUR and w * h > AREA * MIN_PERCENT:
cv.PolyLine(output, [
polygon_points,
], 0, cv.CV_RGB(255, 255, 255), 1, 0, 0)
average_box_area += w * h #acumulacion de totales de areas
bbox_list.append((pt1, pt2)) #lista con todos los bounding box
contour = contour.h_next() #lectura del siguiente contorno, si hay
if len(bbox_list) > 0: #si hubo movimiento
average_box_area = average_box_area / float(
len(bbox_list)) #area promedio de bounding box
new_bbox_list = [
] #nueva lista de bounding box, eliminando los menores al area promedio
for i in range(len(bbox_list)): #recorrido de los bounding box
pt1, pt2 = bbox_list[i] #separacion en dos puntos del bounding box
w, h = abs(pt1[0] - pt2[0]), abs(
pt1[1] - pt2[1]) #obtencion del ancho y largo
if w * h >= average_box_area and w * h > AREA * MIN_PERCENT: #comparacion del area del bounding box con el promedio
new_bbox_list.append(
(pt1,
pt2)) #si es mayor o igual, se queda en la nueva lista
bbox_list = get_collided_bboxes(
new_bbox_list
) #combinacion de varios bounding box en uno si estan en contacto
def __init__(self, bid, contour):
self.bid = int(bid)
self.name = self.label_text_map[bid]
while (ishole(contour)):
contour = contour.h_next()
self.contour = contour
self.area = cv.ContourArea(contour)
self.center_of_mass = im.center_of_mass(contour)
l, t, w, h = cv.BoundingRect(contour)
self.bbox = (l, t, l + w, t + h)
self.width = w
self.height = h
self.near_set = set([])
self.north_set = set([])
self.south_set = set([])
self.east_set = set([])
self.west_set = set([])
def _get_bounding_rects(self, phase, time):
'''
Return rectangle where to look for this marker
@param phase:
@param time:
'''
nrect = []
for cor in self.corners:
(x, y, wx, wy) = cor.get_rectangle(time)
# print (x, y, wx, wy)
nrect.extend([(x, y), (x + wx, y + wy)])
# print nrect
rect = cv.BoundingRect(nrect)
# print rect
return [rect]
def run(self):
while not self._stop.isSet():
task = self.q.get()
if task != None:
obj, image = task
rect = cv.BoundingRect(obj.cont)
siftimage = siftfastpy.Image(rect[2], rect[3])
cv.CvtColor(image, self.gray, cv.CV_BGR2GRAY)
cv.SetZero(self.mask)
cv.FillPoly(self.mask, [obj.cont], cv.Scalar(255))
cv.And(self.gray, self.mask, self.gray)
gnp = np.asarray(cv.GetSubRect(self.gray, rect))
siftimage.SetData(gnp)
t0 = time.time()
# compute keypoints and time how long it takes
frames,desc = siftfastpy.GetKeypoints(siftimage)
self.stats.append((rect[2]*rect[3], time.time() - t0))
# compute feature vector
tmp = np.concatenate((frames[:,2:4], desc), axis=1).astype('float64')
# search in the flann tree for the feature vectors
n = 2
thr = 1.5
result, dists = self.flann.nn_index(tmp, n, checks=32)
# ismatch contains the indices in the testset for which a match is found
ismatch = dists[:,1] > thr * dists[:,0]
# meta contains the index to object-ID mapping
obj.ids = []
for i, res in enumerate(result):
if ismatch[i]:
obj.ids.append(self.meta[res[0]][0])
# obj.ids = [self.meta[res][0] for i, res in enumerate(result) if ismatch[i]]
# transfer keypoints back to full frame coordinates
frames[:,0] += rect[0]
frames[:,1] += rect[1]
obj.frames = frames
obj.desc = desc
def movearea(self, contour, input): points = [] area = 0 while contour: bound_rect = cv.BoundingRect(list(contour)) contour = contour.h_next() # Compute the bounding points to the boxes that will be drawn on the screen pt1 = (bound_rect[0], bound_rect[1]) pt2 = (bound_rect[0] + bound_rect[2], bound_rect[1] + bound_rect[3]) # Add this latest bounding box to the overall area that is being detected as movement area += ((pt2[0] - pt1[0]) * (pt2[1] - pt1[1])) points.append(pt1) points.append(pt2) cv.Rectangle(input, pt1, pt2, cv.CV_RGB(255,0,0), 1) return area
def CenterFunction(R, imghsv):
imgyellowthresh=getthresholdedimgRGeneric(R, imghsv) # creaza mastile de culoare, aici specifice robotului R4
cv.Erode(imgyellowthresh,imgyellowthresh,None,3)#filtru
cv.Dilate(imgyellowthresh,imgyellowthresh,None,6)#filtru
img2=cv.CloneImage(imgyellowthresh)#cloneaza imaginea in img2, useless
storage = cv.CreateMemStorage(0)#creaza un loc in memorie unde sa stocheze, necesar ptr FindContours
contour = cv.FindContours(imgyellowthresh, storage, cv.CV_RETR_CCOMP, cv.CV_CHAIN_APPROX_SIMPLE)#gaseste contururile robotilor
points = []
# This is the new part here. ie Use of cv.BoundingRect()
while contour:
# Draw bounding rectangles
bound_rect = cv.BoundingRect(list(contour)) #creaza un patratel din punctele din contur, ptr afisare/debug
#bound_rect = cv.BoundingRect(contour)
# for more details about cv.BoundingRect,see documentation
pt1 = (bound_rect[0], bound_rect[1]) #
pt2 = (bound_rect[0] + bound_rect[2], bound_rect[1] + bound_rect[3])
points.append(pt1)
points.append(pt2)
cv.Rectangle(color_image, pt1, pt2, cv.CV_RGB(255,0,0), 1)
#pana aici s-a desenat patratul
# Calculating centroids
centroidx=cv.Round((pt1[0]+pt2[0])/2)
centroidy=cv.Round((pt1[1]+pt2[1])/2)
area = cv.ContourArea(list(contour))
#print "CentroidXY:" + str(centroidx) +":" +str(centroidy) + "A:" + str(area)
if(area > 100):
print "CentroidXY:" + str(centroidx) +":" +str(centroidy) + "A:" + str(area)
coords = pack('iiiii', 4,centroidx, centroidy, 0, int(time.time()))
mosq.publish("coords", coords, 0)
contour = contour.h_next()
print contour
# Identifying if blue or yellow blobs and adding centroids to corresponding lists
if (169<cv.Get2D(imghsv,centroidy,centroidx)[0]<180):
red.append((centroidx,centroidy))
elif (100<cv.Get2D(imghsv,centroidy,centroidx)[0]<120):
blue.append((centroidx,centroidy))
elif (67<cv.Get2D(imghsv,centroidy,centroidx)[0]<100):
green.append((centroidx, centroidy))
return
def getpoints(image):
points = []
cv.CvtColor(image, grey, cv.CV_RGB2GRAY)
cv.Threshold(grey, grey, THRESHOLD, 255, cv.CV_THRESH_BINARY)
cv.Copy(grey, grey2)
storage = cv.CreateMemStorage(0)
contour = cv.FindContours(grey, storage, cv.CV_RETR_CCOMP,
cv.CV_CHAIN_APPROX_SIMPLE)
while contour:
bound_rect = cv.BoundingRect(list(contour))
contour = contour.h_next()
pt = (bound_rect[0] + bound_rect[2] / 2,
bound_rect[1] + bound_rect[3] / 2)
#cv.Circle(image, pt, 2, cv.CV_RGB(255,0,0), 1)
points.append(pt)
return points
def run(self):
starttime = time.time()
gray = cv.CreateMat(480, 640, cv.CV_8UC1)
im = cv.LoadImage(self.image)
rect = cv.BoundingRect(self.obj.cont)
siftimage = siftfastpy.Image(rect[2], rect[3])
cv.CvtColor(im, gray, cv.CV_BGR2GRAY)
gnp = np.asarray(cv.GetSubRect(gray, rect))
siftimage.SetData(gnp)
print 'initialization in: %fs' % (time.time() - starttime)
frames, desc = siftfastpy.GetKeypoints(siftimage)
self.obj.frames = frames
self.obj.desc = desc
print '%d keypoints found in %fs' % (frames.shape[0],
time.time() - starttime)
def dewarp(image,window,corners): debug = image.copy() print(corners) # draw red line around edges for debug purposes cv2.polylines(debug, np.int32([[corners[0],corners[1], corners[3],corners[2]]]), True, (0,255,0),7) #show results if DISPLAY: cv2.imshow(window,debug) cv2.cv.ResizeWindow(window,960,640) cv2.waitKey(WAIT_TIME) # Assemble a rotated rectangle out of that info # Todo: move to cV2 np_corners = np.array(corners) rot_box = cv.MinAreaRect2(corners) enc_box = cv.BoundingRect(corners) scaling = 1.0 border = 10 pt_x = enc_box[2]*scaling pt_y = enc_box[3]*scaling new_corners = [(border,border),(pt_x-1+border,border), (border,pt_y-1+border),(pt_x-1+border,pt_y-1+border)] corners = np.array(corners,np.float32) new_corners = np.array(new_corners,np.float32) warp_mat = cv2.getPerspectiveTransform(corners, new_corners) rotated = cv2.warpPerspective(image, warp_mat, (int(round(pt_x+border*2)), int(round(pt_y+border*2)))) #show results if DISPLAY: cv2.imshow(window,rotated) cv2.cv.ResizeWindow(window,960,640) cv2.waitKey(WAIT_TIME) return rotated
def get_rectangle(self, m_d):
'''
Return rectangle where we will be looking for this corner
@param time: current time
@param size: image size
'''
self.get_bounding_points(m_d)
size = m_d.size
(x, y, wx, wy) = cv.BoundingRect(self.bounds)
(x, y, wx,
wy) = x - MIN_SIZE, y - MIN_SIZE, wx + 2 * MIN_SIZE, wy + 2 * MIN_SIZE
min_wx = MIN_RECT * size[0]
min_wy = MIN_RECT * size[1]
if (wx < min_wx):
x -= (min_wx - wx) / 2
wx = min_wx
if (wy < min_wy):
y -= (min_wy - wy) / 2
wy = min_wy
return (x, y, wx, wy)
def getBoundingRects(self):
'''
@return: the bounding boxes of the external contours of the foreground mask.
@note: You must call detect() before getBoundingRects() to see updated results.
'''
#create a list of the top-level contours found in the contours (cv.Seq) structure
rects = []
if len(self._contours) < 1: return (rects)
seq = self._contours
while not (seq == None):
(x, y, w, h) = cv.BoundingRect(seq)
if (cv.ContourArea(seq) > self._minArea):
r = pv.Rect(x, y, w, h)
rects.append(r)
seq = seq.h_next()
if self._filter != None:
rects = self._filter(rects)
return rects
def find_connected_components(frame):
"""Find connected components from an image.
:: iplimage -> [ dict(box<CvBox2D>, rect<CvRect>) ]
Takes as input a grayscale image that should have some blobs in
it. Outputs a data structure containing the 'centers' of the blobs
and minimal rectangles enclosing them. A maximum of 3 blobs are
returned.
The input frame is modified in place.
"""
# A workaround for OpenCV 2.0 crash on receiving a (nearly) black image
nonzero = cv.CountNonZero(frame)
if nonzero < 20:
return []
logging.debug("Segmentation got an image with %d nonzero pixels", nonzero)
contours = get_contours(frame)
#out = draw_contours(frame, contours)
if contours is None:
return []
candidates = []
while contours:
storage = cv.CreateMemStorage(0)
minBox = cv.MinAreaRect2(contours, storage)
boundingRect = cv.BoundingRect(contours, 0)
candidates.append({ 'box' : minBox, 'rect' : boundingRect })
contours = contours.h_next()
candidates = sorted( candidates,
key=lambda x:getArea(x['box']),
reverse=True )
return candidates
def get_top(shape, center, theta):
pt = center
EPSILON = 1.0
angle = theta
scale = 0
(r_x, r_y, r_w, r_h) = cv.BoundingRect(shape)
if SCALE_METHOD == AXIS:
top_pt = center
best_scale = 1
while (pt[0] > r_x and pt[0] < r_x + r_w and pt[1] > r_y
and pt[1] < r_y + r_w):
print
(x, y) = pt
new_x = x + EPSILON * sin(angle)
new_y = y - EPSILON * cos(angle)
pt = (new_x, new_y)
scale += EPSILON
if cv.PointPolygonTest(shape, pt, 0) > 0:
top_pt = pt
best_scale = scale
return (top_pt, best_scale)
else:
return ((r_x + r_w / 2, r_y), max(r_w, r_h))
