Day #212 - OpenCV based Object Tracking Learning's

	#https://docs.opencv.org/3.1.0/db/df8/tutorial_py_meanshift.html
	#Yolo + Meanshift - OpenCV
	#Write comment for each line that you don't understand :)
	import numpy as np
	import cv2
	cap = cv2.VideoCapture(r'E:\Optical_Flow\slow.flv')
	ret, frame = cap.read()
	#frame = cv2.resize(old_frame, (500, 400))
	#set up initial location window
	r,h,c,w = 250,90,400,125 #assign it based on Yolo
	track_window = (c,r,w,h)
	#set up roi for tracking
	roi = frame[r:r+h,c:c+w]
	#Converts an image from one color space to another.
	hsv_roi = cv2.cvtColor(roi,cv2.COLOR_BGR2HSV)
	#The cv2.inRange - three arguments
	#first is the image to perform color detection
	#second - lower limit of the color you want to detect
	#third argument - upper limit of the color you want to detect.
	mask = cv2.inRange(hsv_roi,np.array((0.,60.,32.)),np.array((180.,255.,255.)))
	#cv2.calcHist to calculate the histogram of an image
	#cv2.calcHist(images, channels, mask, bins, ranges)
	#For grayscale images as there's only one channel and [0], [1] or [2]
	#bins - is a list containing the number of bins to use for each channel
	#ranges - is the range of the possibile pixel values which is [0, 256] in case of RGB color space (where 256 is not inclusive).
	roi_hist = cv2.calcHist([hsv_roi],[0],mask,[180],[0,180])
	#its normalized values will be R/S, G/S and B/S (where, S=R+G+B).
	#when normType=NORM_MINMAX (for dense arrays only).
	#The optional mask specifies a sub-array to be normalized.
	#This means that the norm or min-n-max are calculated over the sub-array, and then this sub-array is modified to be normalized
	cv2.normalize(roi_hist,roi_hist,0,255,cv2.NORM_MINMAX)
	#Set up Termination Criteria either 10 iteration or move by atlease 1 pt
	term_crit = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT,10,1)
	while(1):
	ret, frame = cap.read()
	if ret == True:
	hsv = cv2.cvtColor(frame,cv2.COLOR_BGR2HSV)
	#Calculates the back projection of a histogram.
	#images, channels, hist, ranges, scale[, dst]
	dst = cv2.calcBackProject([hsv],[0],roi_hist,[0,180],1)
	#apply mean shift to get the new location
	#move that window to the area of maximum pixel density (or maximum number of points)
	#the movement is reflected in histogram backprojected image.
	#As a result, meanshift algorithm moves our window to the new location with maximum density
	ret, track_window = cv2.meanShift(dst,track_window,term_crit)
	#draw it on image00000000
	x,y,w,h = track_window
	img2 = cv2.rectangle(frame,(x,y),(x+w,y+h),255,2)
	cv2.imshow('img2',img2)
	k = cv2.waitKey(0) & 0xff
	if k==27:
	break
	else:
	break
	cv2.destroyAllWindows()
	cap.release()

view raw py_meanshift.py hosted with ❤ by GitHub

Happy Mastering DL!!!