counting the number of objects in an image using python

Question

I am trying to count the number of objects in this image:

I have a code for that:

import cv2
import numpy as np
image = cv2.imread('d:\obj.jpg')
blurred  = cv2.pyrMeanShiftFiltering(image,31,91) 
gray = cv2.cvtColor(blurred,cv2.COLOR_BGR2GRAY)
ret , threshold = cv2.threshold(gray,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)
cv2.imshow("Threshold",threshold)
_, contours,_=cv2.findContours(threshold,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)
print "Number of contours: %d  "%len(contours)
cv2.drawContours(image,contours,-1,(0,255,255),2)
cv2.namedWindow('Display',cv2.WINDOW_NORMAL)
cv2.imshow('Display',image)
cv2.waitKey()

the number of objects is 9, but the output is 1015.

when I try to show the the objects this is what I get:

How can I fix that? thanks to all :)

I think you're mis-interpreting the meaning of countours: it's probably an image (i.e. an array of values) rather than the number of found objects. — toti08
– toti08, Commented Aug 9, 2018 at 19:47
I’m always baffled at all the problems people try to solve using contours. You don’t need contours in this case. It is cheaper to label the image (see connected component analysis). — Did you display the result of the threshold? Does it actually yield 9 objects? This is one of the nice things about image processing, you can easily examine the result of each step to see if you're headed in the right direction. — Cris Luengo
– Cris Luengo, Commented Aug 9, 2018 at 20:23
@CrisLuengo yes I dispayed the result of threshold, it yields the objects in black dots on a white background — Shlomi Elhaiani
– Shlomi Elhaiani, Commented Aug 9, 2018 at 21:50
Does it show each object as a single dot? I presume not. Also, if the background is white, it will consider the background as the object. You should invert the result of the threshold. — Cris Luengo
– Cris Luengo, Commented Aug 9, 2018 at 22:05

Abhishek Arya · Accepted Answer · 2018-08-09 19:30:46Z

1

You can easily get the area of contours. I would suggest putting up a threshold on area of contours. I mean to say iterate over all the contours and just keep only those which have area greater than a number specified by you and reject others. This way you can avoid small contours which are there due to noises.

answered Aug 9, 2018 at 19:30

Abhishek Arya

5201 gold badge5 silver badges17 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Abhishek Arya Over a year ago

area = cv.contourArea(cnt)

Collectives™ on Stack Overflow

counting the number of objects in an image using python

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related