Font Size: a A A

Research On Active Vision Tracking Technology In Video Surveillance

Posted on:2020-11-13Degree:MasterType:Thesis
Country:ChinaCandidate:N C LiFull Text:PDF
GTID:2428330623459838Subject:Control engineering
Abstract/Summary:PDF Full Text Request
The intelligent security monitoring system is a hot development direction in the security industry,especially the active visual tracking technology in the monitoring system.It has huge application requirements in various scenarios such as schools,banks,and government departments.Major security camera companies have developed intelligent video surveillance systems.In this paper,based on the actual application requirements,a set of gun-ball linkage intelligent video surveillance system is designed and completed.The scene is monitored by the gun-type camera,and the intrusion target is detected in real time.Once the target intrusion is confirmed,the system immediately drives the dome to monitor the target.Outdoor experiments show that the system has good stability,high accuracy and fast response.This article focuses on the image algorithms used in the system.After fully studying the local features of the image,the automatic calibration algorithm of the camera coordinate system and the dome coordinate system is designed.When the camera is fixedly installed,the coordinate relationship can be automatically completed.According to the application requirements of real-time detection,this paper studies the motion detection algorithm.In this paper,the Vibe algorithm is improved,and a semi-dense optical flow estimation algorithm based on daisy feature points is proposed.The experimental results show that: The accuracy of the improved Vibe algorithm is improved,and the detection speed meets the real-time requirements.The performance of the proposed optical flow algorithm is better than the traditional dense optical flow estimation algorithm.In order to improve the motion detection algorithm can not detect the defects of static targets,the KCF tracking algorithm is improved by combining the detection and tracking algorithm,and the color features and scale pool technology are added.The experimental results show that the effect of the combination of detection and tracking algorithm is better than that of motion detection algorithm alone.Finally,the PTZ control algorithm of the camera is studied,and the fuzzy adaptive PID control algorithm is used to control the azimuth of the camera.the outdoor test shows that the linkage tracking function has high accuracy and fast response speed.
Keywords/Search Tags:linkage control, coordinate alignment, target detection, motion tracking
PDF Full Text Request
Related items