You are a guest. Restricted access. Read more.
SCaVis manual

ImRelate Manual and Examples

This manual explains the ImRelate automatic image identification program written in Java. The algorithm does not use pixel fingerprints, so the method can perform image identification even if the original image was re-mastered. The output of this algorithm produces 3 probabilities:

  • Similarity for color intensity (easy part!);
  • Similarity probability for object locations;
  • Similarity probability for object sizes (shape matching);

If the analyzed image is close to the sample image, all three probabilities are close to unity.

The algorithm attempts to detect shapes on the images, and it can be effective even if an image is rotated or resized.

How to compile the package

Sorry, unregistered users have a limited access to the ImRelate manual. This project will soon be integrated with SCaVis. One can unlock ImRelate manual after becoming a full SCaVis member of jWork.ORG. The protected sections explain how to run ImRelate, shows basic examples and explain the library.

How to run examples

Sorry, unregistered users have a limited access to the ImRelate manual. This project will soon be integrated with SCaVis. One can unlock ImRelate manual after becoming a full SCaVis member of jWork.ORG. The protected sections explain how to run ImRelate, shows basic examples and explain the library.

imrelate.txt · Last modified: 2013/12/15 16:44 by admin
CC Attribution-Share Alike 3.0 Unported
Powered by PHP Driven by DokuWiki Recent changes RSS feed Valid CSS Valid XHTML 1.0 Valid HTML5