Approaches to The Development of a Recognition System for Solving the Problem of Determining the Content of Digital Images
Shamiev Muhammadi Olimovich
This article describes a variety of methods for pattern recognition in digital images (image content). The relevance of the development of image content recognition systems is substantiated, as this finds application in most intelligent systems, as well as the essence of the recognition problem. The goals and objectives of the development of image content recognition systems are highlighted. A detailed and generalized classification of digital image processing methods is presented with the most effective ones highlighted.
In particular, the article provides an analysis of the main methods of image segmentation. The content of the most effective and modern methods is disclosed. The analysis of approaches to the construction of recognition systems, the formation of attribute space in the design of complex automated recognition systems with training.
Based on the studies, the authors conclude that the information summarized in the article can serve as a theoretical basis for developing a recognition system in solving the problem of determining the content of digital images. The most effective modern method of text segmentation in an image can be distinguished by a method based on the use of Voronoi diagrams, which can be relied on in the process of developing and creating a system for understanding images, but it, in turn, requires adaptation to modern conditions due to the limited time resource for solving this type of informational tasks.
Volume: Volume 24
Issues: Issue 9
Keywords: computer vision; image content; pattern recognition systems; image segmentation; morphological methods of image analysis; digital image processing methods.