No abstract

No abstract

The paper considers the problem of object detection and recognition, namely, its solution without the use of expensive, resource-intensive and complex data collection and processing systems, with the possibility of its mobility, ease of installation and initial setup. The available ways and methods of solving the problem, their advantages and disadvantages are given. The operation of the system according to an algorithm developed on the basis of the method of object recognition, namely the selection of contours by a filter based on the Canny operator, is presented. The article presents intermediate and final illustrations of the system algorithm at each step, which gives an opportunity to get acquainted with its advantages over classical video surveillance systems and some disadvantages. Elements such as a webcam with a video frame rate of 25 frames per second, a mobile phone and a personal computer with the MatLab2021 programming environment installed (chosen due to the ease of use and the presence of built-in image processing functions) were used to demonstrate the system's performance.

This article introduces the problem of object detection and recognition. The potential mobility of this solution, ease of installation and ease of initial setup, as well as the absence of expensive, resource-intensive and complex image collection and processing systems are presented. Solutions to the problem are demonstrated, along with the advantages and disadvantages of each. The selection of contours by a filter based on the Prewitt operator and a detector of characteristic points is an algorithm of the system, developed within the framework of object recognition techniques. The reader can follow the interim and final demonstrations of the system algorithm in this article to learn about its advantages over traditional video surveillance systems, as well as some of its disadvantages. A webcam with a video frame rate of 25 frames per second, a mobile phone and a PC with the Matlab2020 programming environment installed (due to its convenience and built-in image processing functions) are required to illustrate how the system works.

The problem of step counting has been considered for personal navigation while walking and using mobile sensors with low accuracy. Three primary approaches to computing the acceleration vector magnitude in time domain have enabled step counting. When analyzing data from mobile phone sensors for various pedestrian kinds, environmental factors, and their mobility patterns, several methodologies were compared. Except for the approach of normalized auto-correlation based step counting, which only processes short distances, the walking trajectories have been chosen to be long enough (at least 100 meters) to produce statistically representative results. The specifications for a specific step counting system have been developed.