Research on graphic data formats for compact representation and comparison of images

Abstract

This article investigates common methods of lossy and lossless compression of graphic information. The advantages and disadvantages of compression methods are identified as a result of the research. A comparative analysis of the main capabilities of graphic information compression methods is conducted. The relevance lies in the efficient transmission, processing, and storage of graphic information, as large data volumes require increased network bandwidth and significant resources for data storage. The practical significance lies in solving the task of effectively reducing data sizes by applying well-known compression methods. Based on the study of graphic data formats, the development of algorithms for the computational scheme of "precise" processing of halftone images for pattern recognition is presented. Such a scheme rewrites a multigradient image into a three-level representation and implements a balancing procedure, which allows forming image features in a more compact form and computing the correlation function faster. The effectiveness of using developed methods of compact image representation with correlation comparison of balancing curves is demonstrated compared to traditional correlation comparison of images.