Edge connecting points based image searching and Measurements

摘要

Databases, libraries of digital image and multimedia are considerably expanded in last few years. The growth of a content-based image retrieval (CBIR) system has become main research issue in order to retrieve the desired images successfully and accurately from a large image database. In this paper we concentrated on image features and features extraction and comparisons for similarity. We processed huge amount of images and extracted the each image converted into gray form then extracted edges and calculated connecting points of each image. All connected points we stored into the database then based on query image feature values are compared then images are equating. Observations are showing that this process shown the good result comparatively existed methods.