why I got very low result of IoU when I compared between two images?

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) and I got 0.2822 , 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 help me to fix it please?

I attached the original image (70027-157.jpg) which used in Image segmenter App for labeling, the labeled image is(label7027-157), the model output(new7027-157) which is needed to be evaluated

Answer 1

[Aneela]

Hi @yasmin ismail

The model output, "new7027-157.png" and the groundtruth "BW_groundTruth" are in "grayscale" format not in "binary" format. The "Jaccard Index" needs binary images, the pixel values should be 0 or 1. This difference lead to a low similarity score.

To address the problem, convert both the images into binary masks by applying a threshold before computing IOU. Convert any pixel values greater than "0" to "1" for the foreground, and assign "0" to all other pixels for the background.

Refer to the following MathWorks documentation on "jaccard" for better understanding: https://www.mathworks.com/help/images/ref/jaccard.html