Active Intelligent Monitoring System Based On Two-dimensional Laser Range Finder (LRF) Array

The video surveillance system has always been the mainstream method in the field of surveillance.But it is susceptible to environmental factors such as lighting,the small field of view,the limited monitoring area,and it is difficult to obtain information such as the precise distance and position of the target.The two-dimensional laser range finder(LRF)is less affected by environmental factors.It can get the target image clearly under different lighting conditions.It can directly measure the distance information of the target in high precision(such as millimeter accuracy).In addition,LRF has very large field of view(typically 180 degree and 360 degree field of view).These advantages make it a very well application prospect in monitoring systems.This thesis mainly researches on activity intelligent monitoring system based on LRF Array.Intelligent monitoring of complex environment is achieved through multiple LRFs and active camera.The main work and research contents are as follows:(1)A high-precision calibration of LRF arrays based on virtual dihedron is proposed.In this thesis,the virtual dihedron is constructed by plane calibration of different poses,and the external parameter calibration between multiple LRFs is realized by the geometric mapping relationship of the virtual dihedron.The algorithm only uses the planar template to realize calibration between LRF arrays,then the laser data collected by different LRFs can be mapped into the same reference coordinate system to realize spatial fusion of the laser data.(2)The dynamic pedestrian target detection and tracking algorithm based on two-dimensional laser data is proposed.In this thesis,the Laser Gaussian Model(LGM)is used to model the background of the laser data,so that the dynamic pedestrian target can be detected.Then the DBSCAN(Density-Based Spatial Clustering of Applications with Noise)algorithm is used to cluster the same type of point cloud targets to obtain the refined representation of the pedestrian target.Finally,the pedestrian motion model and Kalman filtering algorithm are used to achieve accurate tracking of pedestrians in different laser fields.(3)A pedestrian target active intelligent sensing system based on LRF array and active camera is proposed.The monitoring area is segmented spatially using the Cell.Thereby,the pedestrian target of the LRF array detection and tracking can be mapped into the corresponding Cell.In addition,according to different Cell,we optimize the PTZ parameters of the active camera to obtain the optimal imaging quality of each Cell,and accurately calibrate the internal and external parameters of the camera in each Cell.Through the cooperative control of the LRF array and the active camera,high-quality pedestrian target imaging is obtained and the matching of pedestrian trajectory and image feature is realized.Thereby,the pedestrian intelligent monitoring is completed.The LRF array-based monitoring system proposed in this thesis makes full use of the advantages of both laser and camera.Through the pedestrian target detection and tracking information acquired by LRF,the optimizing dynamic parameter of the active camera is controlled to achieve optimal imaging of dynamic targets and the matching of pedestrian trajectory and appearance posture characteristics,which provides a certain solution for the intelligent monitoring system.