Camera planning for area surveillance: A new method for coverage inference and optimization using Location-based Service data. (November 2019)

• Record Type: Journal Article
• Authors: Han, Zhigang; Li, Songnian; Cui, Caihui; Song, Hongquan; Kong, Yunfeng; Qin, Fen
• Is Part Of: Computers, environment and urban systems. Volume 78(2019)
• Publication Date: 2019-11
• Subjects: Video surveillance -- Camera planning -- Spatial optimization -- Location-based service -- Maximal coverage location problem; City planning -- Data processing -- Periodicals; Regional planning -- Data processing -- Periodicals; 303.4834
• DOI: 10.1016/j.compenvurbsys.2019.101396
• Languages: English
• ISSNs: 0198-9715

Abstract

Abstract: Along with the rapidly growing volume of public security events, efficient camera planning and configuration methods have been one of the most crucial challenges in the video surveillance field. How to allocate different types of surveillance cameras in an area is one of the fundamental problems; however, limited methods have been available for generating the deployment parameters of cameras. The purpose of the paper is to explore camera planning based on multi-source Location-based Service data. The main idea is to infer the camera coverage by the building footprints, Point of Interests (POI) and social network record (WeChat) data, and to optimize the camera placement using the Maximal Coverage Location Problem-Complementary Coverage (MCLP-CC) model. Based on the probability of cell monitored with the calculation of viewshed analysis, the candidate location with max probability is selected. The essential spots in the surveillance area are uncovered by the combination of the kernel density estimation of POIs and WeChat data. The inference algorithm of the location, the field of view angle, orientation yaw, and visible distance parameters are proposed using the candidate location and critical spots in the viewshed polygon. The MCLP-CC is modeled and implemented by Python scripts and Gurobi software. The experiment shows that the proposed method can generate the detailed camera parameters including location, the field of view angle, orientation yaw, and visible distance with the lower occlusion and overlapping ratio for camera coverage. We believe that the integration of the coverage inference and optimization methods into the existing GIS platform will promote a variety of innovative applications in the camera planning area. Highlights: We introduce a new camera planning method for area surveillance using multi-source Location-based Service data. The detailed camera parameters instead of a single location or fixed type cameras are inferred. The vital spot of an area which is reflected by the static and dynamic objects are utilized in the inference process. It is given rise to the lower occlusion ratio and overlapping ratio with the same coverage ratio.