An Adaptive Method for Dynamic Background Compensation Based on SIFT

Ximing Tao, Qingzhong Jia*, Tingting Du, Hong Huang

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

To address the issue of poor object detection and tracking performance in dynamic scenes, a low-complexity adaptive dynamic background compensation algorithm is proposed to improve the real-time and accuracy of object detection and tracking. The algorithm uses the SIFT feature point matching algorithm to assist in calculating the background motion parameters and introduces an adaptive coefficient with feedback mechanism to reduce the matching calculation range of background motion parameters, thereby achieving dynamic background compensation. While improving the accuracy of background compensation, the algorithm also speeds up the computation to some extent. Experimental results show that this algorithm can improve the accuracy of detecting and tracking moving objects in video sequences with complex dynamic backgrounds while occupying only a small amount of system resources.

源语言英语
主期刊名ITOEC 2023 - IEEE 7th Information Technology and Mechatronics Engineering Conference
编辑Bing Xu, Kefen Mou
出版商Institute of Electrical and Electronics Engineers Inc.
191-197
页数7
ISBN(电子版)9798350334197
DOI
出版状态已出版 - 2023
活动7th IEEE Information Technology and Mechatronics Engineering Conference, ITOEC 2023 - Chongqing, 中国
期限: 15 9月 202317 9月 2023

出版系列

姓名ITOEC 2023 - IEEE 7th Information Technology and Mechatronics Engineering Conference

会议

会议7th IEEE Information Technology and Mechatronics Engineering Conference, ITOEC 2023
国家/地区中国
Chongqing
时期15/09/2317/09/23

指纹

探究 'An Adaptive Method for Dynamic Background Compensation Based on SIFT' 的科研主题。它们共同构成独一无二的指纹。

引用此