Is there a method to capire similarity between two images?

I would like to find method to compaire between two cracked images to measure the similarty, I used jaccard index but not working well , is there another way to measure similarity?

Answer 1

[Image Analyst]

See https://en.wikipedia.org/wiki/Content-based_image_retrieval#Content_comparison_using_image_distance_measures

In MATLAB, you can try dice or ssim:

http://127.0.0.1:64151/static/help/images/obtain-local-structural-similarity-index.html?searchHighlight=ssim&searchResultIndex=10

[yasmin ismail]

@Image Analyst can I use (ssim) instead of IoU to compaire between groundTruth and the output of the model to detect crack ?

A = imread('new7027-157.png');

BW_groundTruth =imread('Label7027_157.png');

[ssimval,ssimmap] = ssim(A,BW_groundTruth);

imshow(ssimmap,[])

title("Local SSIM Map with Global SSIM Value: "+num2str(ssimval))

[Image Analyst]

It is one metric, though it's not usually used on binary images. It's mainly used on gray scale images.

[yasmin ismail]

@Image Analyst why I got very low IoU compaired with ssim

A = logical(imread('new7027-157.png'));

BW_groundTruth =logical(imread('Label7027_157.png'));

similarity = jaccard(A, BW_groundTruth)

similarity =

0.2822

I created GroundTuth by using Image segmenter App and export it to work space , then i used jaccrad index(IoU) , why I got very law result although both images seem same by eye vision, how to fix it, is it related to the way of creating groundtruth or I missed something to do after labeling before calculationg the similarity? can you hel me to fix it please?

I attached the original image (70027-157.jpg) which used in Image segmenter App for labeling

[yasmin ismail]

@Image Analyst