АНАЛІЗ ІСНУЮЧИХ МЕТОДІВ РОЗПІЗНАВАННЯ ОБ'ЄКТІВ НА ПРИКЛАДІ ЗОБРАЖЕННЯХ ВЕБСТОРІНОК

Abstract

Object recognition on web pages is the process of automatic detection and classification of objects on website images. The work analyzes the methods of recognition of objects on digital images for segmentation of websites for further analysis. Based on the analysis of the methods of recognition of objects on the images of web pages, it is concluded that deep learning methods can be very effective for recognizing objects with a lot of data, but they may require significant computing resources and data for training. Geometric functions can be effective for recognizing objects with standardized shapes and sizes, but they can be less effective for recognizing objects with high variability in forms and sizes. The choice of object recognition method on the images of websites depends on various factors, such as the size of the data set, the characteristics of the identified objects, computing resources available for use, and many other factors.