Title :
A method for moving target capture using 3D profile information based on bionic compound eye
Author :
Luo Bo-Ren ; Yan Lei ; Liu Hui-Li ; Sun Hua-Bo
Author_Institution :
Spatial Inf. Integration & Its Applic. Beijing Key Lab., Peking Univ., Beijing, China
Abstract :
Most of the traditional methods of capturing and detecting the moving target are achieved by sequence image difference, which have the shortcomings of small visual field to capture, easy loss of the target and need for high-precision image stabilization. This paper proposed a method based on the bionic compound eye to capture and detect moving targets, which using the control system of bionic compound eye hardware that is designed by the our project team independently to photograph the target area, then generate 3D point clouds of the region, at last, the target can be captured and detected by using appropriate detection rules. This method has been proved to be reasonable and effective.
Keywords :
biocybernetics; image motion analysis; image sequences; object detection; solid modelling; 3D point cloud region; 3D profile information; appropriate detection rules; bionic compound eye hardware control system; high-precision image stabilization; image difference sequence; moving target capture method; moving target detection; bionic compound eye; capture; moving target detection; three-dimensional reconstruction;
Conference_Titel :
Computer Vision in Remote Sensing (CVRS), 2012 International Conference on
Conference_Location :
Xiamen
Print_ISBN :
978-1-4673-1272-1
DOI :
10.1109/CVRS.2012.6421255