comparison of query image feature and 1000 image feature which stored in .mat file

I have stored 1000 image features(16 feature) in v1.mat file and query image feature(16 feature) store in v.mat file how can we compare query image feature(v.mat) with each image feature of(v1.mat) .

[Ameer Hamza]

What is the metric by which you want to compare? What is the size and shape of your matrices?

[vijaylakshmi sajwan]

size of the first matric is 1*16 and size of the second matric is 1000*16

[Image Analyst]

What does compare mean to you? Do you have a class for each of the 1000 images, and want to find out which class the query image belongs to using some classification method like KNN or a decision tree or something?

[vijaylakshmi sajwan]

compare mean to find the minimum distance between query image feature and database images feature and 5 least distance images will display

Answer 1

[Ameer Hamza]

Try this,

V1 = rand(1000,16); % random values for example
V = rand(1,16);

[~,idx] = min(sum((V1-V).^2, 2));
closest_sample = V1(idx,:);

It finds a sample from v1 closest to v, based on Euclidean distance.

[vijaylakshmi sajwan]

sir i have two mat file one is v1.mat file (that store 1000 image feature) and v.mat file (that store query image feature). both are different file. but now i want to use eucledian distance matrix find out the minimum distance images.

[Ameer Hamza]

Yes, the above code shows how to do that. Is there any issue when this method?