def process(filename):
fs = fr.read_frames(filename)
kernel = np.ones((3,3), np.uint8)
newFs = []
for f in fs:
f = cv2.resize(f, None, fx=0.4, fy=0.4)
f = cv2.equalizeHist(f)
f1 = f.copy()
# f = cv2.medianBlur(f, 3)
# cnts = cd.contours(f)
# cv2.drawContours(f, cnts, -1, 255)
f[f < f.max()-10] = 0
f = cv2.morphologyEx(f, cv2.MORPH_OPEN, kernel)
# show_frame(f)
newFs.append(f)
# params for ShiTomasi corner detection
feature_params = dict( maxCorners = 100,
qualityLevel = 0.3,
minDistance = 7,
blockSize = 7 )
# Parameters for lucas kanade optical flow
lk_params = dict( winSize = (15,15),
maxLevel = 2,
criteria = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03))
p0 = cv2.goodFeaturesToTrack(newFs[0], mask = None, **feature_params)
# Create a mask image for drawing purposes
mask = np.zeros_like(newFs[0])
# Create some random colors
color = np.random.randint(0,255,(100,3))
for fo, fn in zip(newFs, newFs[1:]):
# calculate optical flow
p1, st, err = cv2.calcOpticalFlowPyrLK(fo, fn, p0, None, **lk_params)
# Select good points
good_new = p1[st==1]
good_old = p0[st==1]
# draw the tracks
for i,(new,old) in enumerate(zip(good_new,good_old)):
a,b = new.ravel()
c,d = old.ravel()
cv2.line(mask, (a,b),(c,d), color[i].tolist(), 2)
cv2.circle(fn,(a,b),5,color[i].tolist(),-1)
fn = cv2.add(fn, mask)
show_frame(fn)
# Now update the previous frame and previous points
fo = fn.copy()
p0 = good_new.reshape(-1,1,2)
def init():
global axes_, lines_
global cap_, args_
global data_
global frames_
global stride_
global t_, y_
global fps_
videoFile = args_['video']
print('[INFO] Processing %s' % videoFile )
frames_ = fr.read_frames( videoFile )
fps_ = args_['fps']
tmax = len(frames_) / float(fps_)
stride_ = len(y_) / len(frames_)
print('[INFO] Total frames: %s' % len(frames_))
print('[INFO] Computed fps: %s' % fps_)
print('[INFO] Stride to take: %s' % stride_)
return lines_.values()
def init():
global axes_, lines_
global cap_, args_
global data_
global frames_
global stride_
global t_, y_
global fps_
videoFile = args_['video']
frames_ = fr.read_frames( videoFile )
data_ = scipy.io.loadmat( args_['activity'] )
data_ = data_['final']
t_, y_ = data_[:,0], data_[:,1]
fps_ = args_['fps']
tmax = len(frames_) / float(fps_)
stride_ = len(y_) / len(frames_)
print('[INFO] Total frames: %s' % len(frames_))
print('[INFO] Computed fps: %s' % fps_)
print('[INFO] Stride to take: %s' % stride_)
axes_['activity'].set_xlim([0, tmax])
axes_['activity'].set_ylim([y_.min() - 2, y_.max() + 2])
return lines_.values()
def main( inputfile, **kwargs):
logger.info('Processing %s ' % inputfile)
frames = fr.read_frames( inputfile, 10 )
fShape = frames[0].shape
logger.info('Got total %s frames' % len(frames))
print("Done reading frames")
data = np.dstack( frames )
rows, cols = data.shape[0], data.shape[1]
pixals = []
for r in range(rows):
for c in range(cols):
vec = data[r, c, :]
pixals.append( Pixal(vec, (r,c)) )
for p in pixals:
add_pixal( p )
print("Total nice pixals: %s" % len(pixal_of_interest_))
img = np.zeros( shape = fShape )
for loc in pixal_of_interest_:
img[loc] = pixal_of_interest_[loc].max
pylab.imshow( img )
pylab.show()
def main(inputfile, **kwargs):
logger.info('Processing %s ' % inputfile)
frames = fr.read_frames(inputfile, 10)
fShape = frames[0].shape
logger.info('Got total %s frames' % len(frames))
print("Done reading frames")
data = np.dstack(frames)
rows, cols = data.shape[0], data.shape[1]
pixals = []
for r in range(rows):
for c in range(cols):
vec = data[r, c, :]
pixals.append(Pixal(vec, (r, c)))
for p in pixals:
add_pixal(p)
print("Total nice pixals: %s" % len(pixal_of_interest_))
img = np.zeros(shape=fShape)
for loc in pixal_of_interest_:
img[loc] = pixal_of_interest_[loc].max
pylab.imshow(img)
pylab.show()