def run(self):
hist = cv2.createHist([180], cv2.CV_HIST_ARRAY, [(0,180)], 1 )
backproject_mode = True
while True:
frame = cv2.QueryFrame( self.capture )
# Convert to HSV and keep the hue
hsv = cv2.createImage(cv2.GetSize(frame), 8, 3)
cv2.cvtColor(frame, hsv, cv2.CV_BGR2HSV)
self.hue = cv2.createImage(cv2.GetSize(frame), 8, 1)
cv2.split(hsv, self.hue, None, None, None)
# Compute back projection
backproject = cv2.createImage(cv2.GetSize(frame), 8, 1)
cv2.calcArrBackProject( [self.hue], backproject, hist )
# Run the cam-shift (if the a window is set and != 0)
if self.track_window and is_rect_nonzero(self.track_window):
crit = ( cv2.CV_TERMCRIT_EPS | cv2.CV_TERMCRIT_ITER, 10, 1)
(iters, (area, value, rect), track_box) = cv2.camShift(backproject, self.track_window, crit) #Call the camshift !!
self.track_window = rect #Put the current rectangle as the tracked area
# If mouse is pressed, highlight the current selected rectangle and recompute histogram
if self.drag_start and is_rect_nonzero(self.selection):
sub = cv2.getSubRect(frame, self.selection) #Get specified area
#Make the effect of background shadow when selecting a window
save = cv2.cloneMat(sub)
cv2.convertScale(frame, frame, 0.5)
cv2.copy(save, sub)
#Draw temporary rectangle
x,y,w,h = self.selection
cv2.rectangle(frame, (x,y), (x+w,y+h), (255,255,255))
#Take the same area but in hue image to calculate histogram
sel = cv2.getSubRect(self.hue, self.selection )
cv2.calcArrHist( [sel], hist, 0)
#Used to rescale the histogram with the max value (to draw it later on)
(_, max_val, _, _) = cv2.getMinMaxHistValue( hist)
if max_val != 0:
cv2.convertScale(hist.bins, hist.bins, 255. / max_val)
elif self.track_window and is_rect_nonzero(self.track_window): #If window set draw an elipseBox
cv2.ellipseBox( frame, track_box, cv2.CV_RGB(255,0,0), 3, cv2.CV_AA, 0 )
cv2.showImage( "CamShiftDemo", frame )
cv2.showImage( "Backprojection", backproject)
cv2.showImage( "Histogram", self.hue_histogram_as_image(hist))
c = cv2.waitKey(7) % 0x100
if c == 27:
break
def dilate(ary, N, iterations):
kernel = np.zeros((N, N), dtype=np.uint8)
kernel[(N - 1) / 2, :] = 1
dilated_image = cv2.dilate(ary / 255, kernel, iterations=iterations)
kernel = np.zeros((N, N), dtype=np.uint8)
kernel[:, (N - 1) / 2] = 1
dilated_image = cv2.dilate(dilated_image, kernel, iterations=iterations)
dilated_image = cv2.convertScale(dilated_image)
return dilated_image
def hue_histogram_as_image(self, hist):
""" Returns a nice representation of a hue histogram """
histimg_hsv = cv2.createImage((320, 200), 8, 3)
mybins = cv2.cloneMatND(hist.bins)
cv2.log(mybins, mybins)
(_, hi, _, _) = cv2.minMaxLoc(mybins)
cv2.convertScale(mybins, mybins, 255. / hi)
w,h = cv2.getSize(histimg_hsv)
hdims = cv2.getDims(mybins)[0]
for x in range(w):
xh = (180 * x) / (w - 1) # hue sweeps from 0-180 across the image
val = int(mybins[int(hdims * x / w)] * h / 255)
cv2.rectangle(histimg_hsv, (x, 0), (x, h-val), (xh,255,64), -1)
cv2.rectangle(histimg_hsv, (x, h-val), (x, h), (xh,255,255), -1)
histimg = cv2.cvtColor(histimg_hsv, cv2.COLOR_HSV2BGR)
return histimg
def hue_histogram_as_image(self, hist):
""" Returns a nice representation of a hue histogram """
histimg_hsv = cv2.createImage( (320,200), 8, 3)
mybins = cv2.cloneMatND(hist.bins) #Contain all values
cv2.log(mybins, mybins) #Calculate logarithm of all values (so there are all above 0)
(_, hi, _, _) = cv2.MinMaxLoc(mybins)
cv2.convertScale(mybins, mybins, 255. / hi) #Rescale all element to get the highest at 255
w,h = cv2.getSize(histimg_hsv)
hdims = cv2.getDims(mybins)[0]
for x in range(w):
xh = (180 * x) / (w - 1) # hue sweeps from 0-180 across the image
val = int(mybins[int(hdims * x / w)] * h / 255)
cv2.rectangle( histimg_hsv, (x, 0), (x, h-val), (xh,255,64), -1)
cv2.rectangle( histimg_hsv, (x, h-val), (x, h), (xh,255,255), -1)
histimg = cv2.createImage( (320,200), 8, 3) #Convert image from hsv to RGB
cv2.cvtColor(histimg_hsv, histimg, cv2.CV_HSV2BGR)
return histimg
def float_version(img):
return cv2.convertScale(img, 1 / 255.0)
while True:
_, color_image = capture.read()
color_image = cv2.GaussianBlur(color_image, (21, 21), 0)
########################
if not difference: #For the first time put values in difference, temp and moving_average
difference = color_image.copy()
temp = color_image.copy()
cv2.convertScaleAbs(color_image, moving_average, 1.0, 0.0)
else:
cv2.runningAvg(color_image, moving_average, 0.020,
None) #Compute the average
# Convert the scale of the moving average.
cv2.convertScale(moving_average, temp, 1.0, 0.0)
# Minus the current frame from the moving average.
cv2.absDiff(color_image, temp, difference)
#Convert the image so that it can be thresholded
cv2.cvtColor(difference, grey_image, cv2.COLOR_RGB2GRAY)
cv2.threshold(grey_image, grey_image, 70, 255, cv2.THRESH_BINARY)
cv2.dilate(grey_image, grey_image, None, 18) #to get object blobs
cv2.erode(grey_image, grey_image, None, 10)
# Find contours
storage = cv2.createMemStorage(0)
contours = cv2.findContours(grey_image, storage, cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
