|
ИСТИНА |
Войти в систему Регистрация |
ИПМех РАН |
||
Visual surveilance system with target-oriented distorsions and segmentation algorithmsстатья
- Авторы: Belchenko Ph M., Ermolov I.L., Ostrikov P.P.
- Журнал: Известия Волгоградского государственного технического университета
- Номер: 4
- Год издания: 2023
- Издательство: ВолгГТУ
- Местоположение издательства: Волгоград
- Первая страница: 18
- Последняя страница: 22
- DOI: 10.35211/1990-5297-2023-4-275-18-22
- Аннотация: This paper studies a new approach to improve efficiency of modern surveillance and telemetry systems in robotics. It is based on theory of perceptive space applied to visual surveying of critical objects. A main goal of the study is to improve efficiency of human operator while inspecting or diagnosing various emerging situations. As a key solution we propose to introduce target distortions into images received from surveillance cameras. Additionally segmentation algorithms for peculiar situations also are presented. Results of final of a robot testing of joint performance of segmentation combined with mean shift after perception transformation are discussed.Part of this work was supported by the Ministry of Science and Higher Education within the framework of the Russian State Assignment under contract No. 123021700055-6
- Добавил в систему: Бельченко Филипп Михайлович