The System Guided By Machine Vision Based On Improved Triangulation Method

The industrial robot has a very wide application in all aspects of production and processing since its strong flexibility and precise operation trajectory.Today,most of the robots applied in industrial applications preset trajectory to do the production and processing of artifacts.This kind of method has various limitation.Once the position of the artifact is opposite to the preset position,this kind of operation mode that carries out the preset track will not satisfy the actual requests.Taking into account the special advantages of machine vision technology,with high flexibility and operation accuracy.The application of machine vision technology in the field of industrial production has made great progress.The main research contents of this paper is how to guide the control system to robot vision technology.Then the collected visual information works in image processing,image target detection,and feature extraction.The 3D feature information extracted of the object represents the position signal of posture adjustment and the sequence of processing on the mechanical arm in the follow-up process.The advantages of this system are as follow-ups.It will improve the traditional robot control system,which is impossible to adjust according to the operating environment and real-time processing path of artifact position.The traditional robot control system can only repeat the planning path mechanically.The improved system combines the advantages of visual guidance technology and robot control system,which represents the formation of a more rapid and effective system with high accuracy.The real-time system provides a more flexible processing mode that can be adapted according to different types of artifacts and the position as well as the attitude of industrial robot operation,with more flexible application and wide application scope.The ideas of designing the improved system and innovative methods of application are as follows.(1)The application of the theory that has been widely used in image processing and machine vision to design the algorithm for acquiring the position and attitude information of the artifact.At first the application of improved triangulation method in the stereo vision system using a binocular camera and a cross semiconductor laser,is getting the depth information of the target artifact plane and the tilt angle obtained by the laser line length.Second the binocular camera is designed for gaining the target plane coordinates,which is obtained by the transformation of coordinates from the center of the characteristic circle for the artifact pixel feature.Then the rotation angle of the artifact in the target plane can be achievable by transforming from the tilt angle of the feature line on the image.The information are integrated together at the end.The algorithm for obtaining all the necessary information about the position and attitude of the target artifact comes true.(2)The calibration of the improved laser triangulation measurement system is designed.First,the internal and external parameters using standard checkerboard of binocular camera calibration obtain binocular camera internal and external parameters.Then the cross semiconductor laser scattering angle is calibrated,thus we get the numerical scattering angle as the data support for the follow-up depth information and numerical solution of tilt angle information at the same time.(3)The software and hardware selection of the system guided by machine vision are achievable.First,the actual selection of hardware for the corresponding function module,including the selection of cross semiconductor laser,binocular camera,manipulator and control system form a complete hardware system designed meticulously.The selection of the hardware is proceed in sequence.Then the next step is the selection of the software.(4)The feature detection and the corresponding information extraction algorithm come true by using the existing function of the design and implementation of the process.The process includes a series of preconditioning such as binary processing and filtering.Then the application of target detection and image recognition technology can measure the 3D space position and attitude angle of the target artifact to be processed.This method gets the source control information of the system.Then the complete information needed can be seen in the paper.In this paper,the analysis and the comparing of the results provide evidence to verify the actual effect of the system.The application of special position coordinates and attitude angle provide evidence to validate the correctness of the algorithm above.Then the conclusion of the paper provides verification by testing the general attitude.There are still many deficiencies in this paper,which will provide suggestions for further improvement of the work.