.png){kind=icon}
.png){kind=icon}
Return
Visual detection and tracking algorithms for human motion
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-023-15231-1.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In dense scenes, a large number of individuals can introduce serious complications for motion detection, such as blurred vision, chaotic scenes, and complex behaviours. For low-density pedestrian detection and tracking algorithms, the accuracy is greatly reduced for both detection and tracking. High-density detection or tracking fails too when these problems are encountered in high-density scenes. In light of the above problems, a detection algorithm and a tracking algorithm based on the human head and shoulder model are proposed. A support vector machine is used to train the classifier by machine learning. The detection algorithm proposed in this paper achieves a detection accuracy of 94% by using the MIT and INRIA datasets. The average accuracy of pedestrian tracking in high-density scenes is approximately 95%.
Keywords:
Pedestrian detection
Pedestrian tracking
Particle filter
Robustness
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W