Depth measurement based on omni-directional camera and pixel offset

Abstract

3D depth measurement is important in robot navigation, remote manipulation and auto-driving system etc. It can be used to avoid obstacles and identify objects. Using omni-directional camera and infrared point cloud structured light, this paper presents a 3D depth measurement method based on pixel offset theory. Firstly, based on triangulation, the pixel offset values between the reference and object images can be calculated. Secondly, the 3D depth data can be measured and used to generate the color depth image. Finally, to compare with actual measured values, in order to verify the effectiveness of the method. Experimental results show that the reconstructed depth data have a good correlation with the actual distance.