Design And Implementation Of Target Tracking System Base On Fusion WIFI And Visual Information

In recent years,surveillance systems have been widely popularized.The construction and application of surveillance systems are playing an increasingly important role in combating crime and maintaining stability.Video reconnaissance has become a new method for public security organs to investigate and solve cases.In video detection applications,tracking and retrieval of specific targets are mostly done manually,which consumes a lot of manpower,material resources,and time,which affects the efficiency of crime detection.In order to solve the above problems,the pedestrian detection and tracking and pedestrian re-recognition technology based on computer vision has become a research hotspot,but it is limited by the single visual information.Strong,high computing power requirements.However,with the popularity of smart wireless terminals and WIFI,pedestrian positioning and tracking using WIFI information has become possible.Pedestrian positioning and tracking based on wireless information has the advantages of low computational complexity and strong achievability,but wireless positioning technology is subject to multipath The effects of propagation and non-line-of-sight interference have the disadvantage of poor positioning accuracy.This system has developed a non-cooperative wireless information collection device by using WIFI information instead of image information as the basis for pedestrian identity recognition through understanding the needs of the target tracking system and technical research.This system uses pedestrian detection and tracking technology based on YOLOv3 to visually track pedestrians in the surveillance area.By deploying wireless information collection equipment in the video surveillance area,information about the communication frames of pedestrians’ wireless devices and wireless access points is obtained.The MAC address of the wireless device is used as the identification feature of pedestrian identity.The signal strength measured by the collection device is used to wirelessly locate the pedestrian.According to the positioning result,the MAC address is assigned to the pedestrian detection frame,which matches the visual information of the pedestrian and the WIFI information.And based on this fusion tracking method,a target tracking system that combines WIFI information and visual information is designed and realized.This thesis first analyzes the research background and development trend of the target tracking system,introduces the related technologies and concepts of wireless network,pedestrian detection tracking and system development,and secondly takes the software development process as the logic of writing,elaborating the functional requirements and Functional requirements,the design of the system’s overall architecture,then introduced the system’s detailed design,implementation process and effect,and finally through the system test,completed the target tracking system that basically meets the actual needs.