how to find the eucledian distance
2 views (last 30 days)
Show older comments
i have 10 gray images of size [256 256],ihave extracted the feature vector of all images.the fv of each image is of size[1 65536].then i have stacked all 10 fv into one matrix.now d size of this single matrix is [10 65536].i have taken the 1st image as the query image,now i want to find the eucledian distance between the query image and all the other 10 image (including query image ,it will be 0 anyway).i need the distance matrix of size 10 rows and 1 column...someone please help
0 Comments
Answers (1)
Image Analyst
on 2 Oct 2012
Huh? 65536 features? Usually you have only a handful of features per image. It looks like you're saying that every single pixel in the image is a feature. Comparing every pixel in the image to every pixel in some other image is probably not a good method of comparing the images unless you're trying to find images that are really close, like one that has undergone compression and decompression to the original. In that case you can check out the PSNR or supposedly better, the SSIM.
Wayne King says "measerr() in the Wavelet Toolbox computes PSNR and a few other approximation quality measures."
0 Comments
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!