3D coordinate identification of object using binocular vision system for mobile robot

Abstract

This paper aims to develop an intelligent robot binocular vision system using image processing principle and edge of the detection algorithms can be clearly presented object contours. Utilizing random cyclometry with Hough transform random lines can find a specified object. We can explore the image recognition application technology from this study. In this paper, with the binocular vision system coordinates using the service robot localization algorithm the target can be captured quickly and accurately. Through two Webcam cameras calibration, we can find the internal and external parameters. Then the 3D space coordinates of fixed position of the target object can be calculated using the principle of binocular vision trigonometry. The distance between the target and the robot can be found through the coordinate transformation systems. Through the triangular geometry theoretical derivation and verification, the system error can be within a reasonable range. It is also confirmed that the proposed method can achieve the desired accuracy and reliability.