DEV Community

Cover image for How to count objects on an image with Python
Stokry
Stokry

Posted on

How to count objects on an image with Python

In this tutorial, you will learn how you can count the number of objects on an image with Python using CV2.

This is our test image:

enter image description here

Let's jump to the code:

First we need to import our dependencies:



import cv2
import numpy as np


Enter fullscreen mode Exit fullscreen mode

First we need to read our image:



img = cv2.imread('test.jpg')


Enter fullscreen mode Exit fullscreen mode

then we will be converting it into grayscale



img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)


Enter fullscreen mode Exit fullscreen mode

after that, we doing thresholding on image



_, thresh = cv2.threshold(img, 225, 255, cv2.THRESH_BINARY_INV)
kernal = np.ones((2, 2), np.uint8)


Enter fullscreen mode Exit fullscreen mode

then we are doing dilation process, removing black distortion:



dilation = cv2.dilate(thresh, kernal, iterations=2)


Enter fullscreen mode Exit fullscreen mode

next step is finding contour shapes:



contours, hierarchy = cv2.findContours(
    dilation, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)


Enter fullscreen mode Exit fullscreen mode

Then we are getting number of contours (objects found):



objects = str(len(contours))


Enter fullscreen mode Exit fullscreen mode

We can now print number of objects on an image



text = "Obj:"+str(objects)
cv2.putText(dilation, text, (10, 25),  cv2.FONT_HERSHEY_SIMPLEX,
            0.4, (240, 0, 159), 1)


Enter fullscreen mode Exit fullscreen mode

For the lasr step we can show, original, threshold and dilation image:



cv2.imshow('Original', img)
cv2.imshow('Thresh', thresh)
cv2.imshow('Dilation', dilation)

cv2.waitKey(0)
cv2.destroyAllWindows()


Enter fullscreen mode Exit fullscreen mode

This is our final result:

enter image description here

Thank you all.

Top comments (2)

Collapse
 
mccurcio profile image
Matt Curcio

Nice step-by-step explanations.
Good job

Collapse
 
stokry profile image
Stokry

Thanks😀