DEVELOPMENT AND STUDY OF PERFORMANCE IN RELATION TO MOBILE COMPUTER VISION MODULE

Abstract

In the current era of rapid IT development and computerization of society, highly effective systems for large data analysis play an important role in various spheres of human life. The constantly-growing volume of data, as well as its rapid modification in order to improve the efficiency of use, create the necessity for the modernization of existing methods and the development of new approaches to data processing. In this paper, we study algorithms and methods of concurrent programming and load distribution with the maximum use of available computing resources. In this work, in accordance with various design architectures of convolutional neural networks, we prepared the software for image recognition tasks with load distribution of the cluster system for the single-board computers "Raspberry pi". In this article, we describe software testing in various situations.