Binocular Stereo Vision System Camera Calibration Method And Control System

Robot vision is one of the key technologies for intelligent robots,which plays a decisive role for intelligent robot. Visual servo is the automatic acquisition and analysis of images for the purpose of control the robot. So after obtained the direct images feedback information, visual servo is the fast image processing and gives feedback information in the short time, which constitutes the robot's servo control. Robot visual servo system involves a wide range of content, for example image processing, kinematics, dynamics, control theory, real-time calculation and so on. In this paper, we study and discuss the two aspects of the visual servo system control structure and stereo vision calibration.Through analysis and research the visual calibration models and methods at home and abroad, we summarized the key technology, which need to be solved during the progress of studying the machine vision. We expounded the principle of binocular stereo vision, proposed the solution of the camera axis perpendicular to the target surface in the calibration model after analyzing and researching the traditional calibration methods. After analyzing the characteristics round projection model, we show the rule that the projection was elliptic when the CCD surface was not parallel with the target surface. We also put forward a new camera calibration method based on the feature circular, which can simplify the calibration process.Described the composition theory of binocular vision control system, we designed the binocular visual servo control system's hardware and software after determining the overall program of the binocular stereo vision system. The hardware components of the binocular vision control system comprise mechanical structure, stepper motor drive hardware, camera links and the connection of the control system. Binocular vision control system software part includes the control of the mechanical structure, image acquisition, image processing and so on.After established camera calibration system and analysis the control process of the calibration, we developed the judging criteria of the target surface perpendicular to the platform and the camera optical axis perpendicular to the target surface and analyzed the system's adjustment process. On this basis, we calculated the camera's focal length and target depth information in the calibration test. Through the test, calibration process become very simple and we also obtained the final ideal calibration results.