Thursday, December 26, 2024
Google search engine
HomeLanguagesMeasure similarity between images using Python-OpenCV

Measure similarity between images using Python-OpenCV

Prerequisites: Python OpenCV
Suppose we have two data images and a test image. Let’s find out which data image is more similar to the test image using python and OpenCV library in Python.
Let’s first load the image and find out the histogram of images.
Importing library 
 

import cv2

Importing image data 
 

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

Converting to gray image 
 

gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

Finding Histogram 
 

histogram = cv2.calcHist([gray_image], [0], 
                              None, [256], [0, 256])

Example:
Images used:
data1.jpg 
 

data2.jpg 
 

test.jpg 
 

 

Python3




import cv2
   
      
# test image
image = cv2.imread('cat.jpg')
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
histogram = cv2.calcHist([gray_image], [0], 
                         None, [256], [0, 256])
   
# data1 image
image = cv2.imread('cat.jpeg')
gray_image1 = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
histogram1 = cv2.calcHist([gray_image1], [0], 
                          None, [256], [0, 256])
   
# data2 image
image = cv2.imread('food.jpeg')
gray_image2 = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
histogram2 = cv2.calcHist([gray_image2], [0], 
                          None, [256], [0, 256])
   
   
c1, c2 = 0, 0
   
# Euclidean Distance between data1 and test
i = 0
while i<len(histogram) and i<len(histogram1):
    c1+=(histogram[i]-histogram1[i])**2
    i+= 1
c1 = c1**(1 / 2)
   
  
# Euclidean Distance between data2 and test
i = 0
while i<len(histogram) and i<len(histogram2):
    c2+=(histogram[i]-histogram2[i])**2
    i+= 1
c2 = c2**(1 / 2)
   
if(c1<c2):
    print("data1.jpg is more similar to test.jpg as compare to data2.jpg")
else:
    print("data2.jpg is more similar to test.jpg as compare to data1.jpg")


Output : 
 

data1.jpg is more similar to test.jpg as compare to data2.jpg

 

RELATED ARTICLES

Most Popular

Recent Comments