计算机视觉
计算机科学
卡尔曼滤波器
人工智能
帧(网络)
跟踪(教育)
车辆跟踪系统
职位(财务)
钥匙(锁)
运动(物理)
跟踪系统
智能交通系统
工程类
电信
心理学
教育学
土木工程
计算机安全
财务
经济
作者
Bingqing Niu,Hongtao Wu,Ying Meng,Xiao Han,Junyi Ren
标识
DOI:10.1109/icivc58118.2023.10270036
摘要
With the establishment and improvement of highway video surveillance system, traffic target recognition has become a key concern technology in the field of transportation. Aiming at the problem that the existing traffic target recognition technology has poor tracking effect on obscured and blurred targets in continuous motion vehicles, an improved Camshift motion vehicle tracking algorithm is proposed, which predicts the possible position of the obscured target in the current frame by Kalman filtering algorithm, and searches for the target in the position neighborhood of the image frame by Camshift algorithm using color information, combining with the multi-vehicle Tracking chain principle to achieve multi-vehicle tracking. The experiments show that this method can track the moving vehicles effectively in real time, and even when the vehicles are obscured in large proportion, the vehicles can still be tracked correctly, which verifies the effectiveness of this method.
科研通智能强力驱动
Strongly Powered by AbleSci AI