Query image retrieval using similarity measure classifier

Abstract EN

Query by image content (QBIC) is the application of computer vision to the image retrieval problem, that is, the problem of searching for digital images in large databases.

Without the ability to examine image content, searches must rely on metadata such as captions or keywords, which may be laborious or expensive to produce.

In this paper, we work on specifi image to retrieve images that are similar in shape, texture and/or color to an image using some similarity measures.

We show statistical results of the different techniques that we use in different stages of our QBIC system.

We tested five different Sigma in GRNN Neural Network and five different image enhancement techniques.

Also, we have six different gradient masks, so we specify which enhancement technique works better with which masks, to retrieve the image in high value of similar measure.