Checking Human Movement in Computer Vision Systems Based on the Skeleton Model

Beknazarova, S.S., Jaumitbaeva, M.K.

Рақамли технологияларнинг назарий ва амалий масалалари · 2024-yil

Annotatsiya

Analysis of human movement in computer vision systems is carried out in two-dimensional and three-dimensional spaces. A distinctive feature of image analysis in two-dimensional space is the use of images from a single camera and/or multiple body sensors. Such an approach leads to a rapid accumulation of error and, as a result, a low resolution of the image. 3D analysis uses multiple cameras, and images are represented as sets of volumetric elements. Despite the high accuracy of three-dimensional analysis, it is associated with high computational complexity and load on the internal computer network. The aim of the study is to develop a model using a single camera image. However, the goal is to achieve an accuracy close to the three-dimensional analysis. In the considered model, the skeleton represented as a human figure is calculated based on the depicted asegmatic graph. Analysis structures of human figures were performed and fifteen main points were highlighted. Mathematical description of physical and logical connections is given and researched. The speed and spatial characteristics of points and connections describe the general dynamics of movement. A description of the human motion model and its construction model for specific images are presented. Based on the proposed model, an algorithm for collecting and analyzing information about the relative location and speed characteristics of graph nodes and edges has been developed. The developed model can be used to collect information about the standard (normal) dynamics of human movement. If a dynamic deviating from the standard is detected, the behavior is defined as deviant. The proposed algorithm can be used in computer vision systems to detect and analyze human movements, as well as to detect movement deviations from the reference.

Maqola ma’lumotlari
MualliflarBeknazarova, S.S., Jaumitbaeva, M.K.
JurnalРақамли технологияларнинг назарий ва амалий масалалари
Nashr sanasi2024-04-09
Jild7
Son1
Betlar39-45
TilRus
DOI10.62132/ijdt.v7i1.159

Kalit so‘zlar

computer vision, human motion analysis, behavior analysis, motion detection, skeletal model, kompyuter ko‘rish, inson harakatini tahlil qilish, xatti-harakatlar tahlili, harakatni aniqlash, skelet modeli

Ilmiy soha

Рақамли технологияларнинг назарий ва амалий масалалари jurnalidan boshqa maqolalar

Рақамли технологияларнинг назарий ва амалий масалалари — barcha maqolalar