Research On Billet Position Detection And Image Processing Technology

With the continuous development of computer hardware and software technology and machine vision field,the application of the position detection system based on visual processing is widely used in automatic production.For cooperation with Beijing science and technology enterprise development of the marking device is in the process of spraying the billet;the different position difference arrangement of the billet will lead to the collision of the gun and the incomplete of the spray font.This paper studies and discusses the combination of image processing and analysis techniques to measure the spatial position of the steel billet,the main work of this paper is as follows:1.Introduce the present situation of domestic and foreign research on the field of the subject,and then analyze the advantages of the use of Kinect sensor as the hardware acquisition device of the whole system.Through the analysis of the Kinect structure and principle of the depth data measurement,this paper use the C++ programming to achieve the depth and color information.2.The algorithm of digital image processing in the system is designed in detailed,including the registration of the Kinect depth and color information.Through the comparison of the filtering effect of a variety of filters,the median filter can filter out the interference noise and keep the details of the image preserved.Comparison of the detection results of different edge detection operators,adopt can precisely get single primality edge and strong anti-interference ability of Canny operator.Finally by comparing the two kinds of expression of image contour,use the method of vertex sequence and fitting the contour with maximum boundary then the contour information of the target is obtained and stored.3.The basic principle of the camera calibration algorithm is studied based on the analysis of the coordinate transformation relationship in the process of camera imaging.Analysis of the imaging process three sources of error and the effective method to reduce the error is proposed.Then the internal and external parameters of the Kinect are calibrated.The transformation model of image pixel coordinates to the thee-dimensional coordinate information is established and the 3D spatial information of the target is obtained.The paper expounds the reason of error depth measurement and through the servo positioning device its accuracy is analyzed,then the error distribution of the measurement value is obtained.Combined with the practical application of this paper,the measurement range of Kinect si determined and the error is corrected.Finally the system designed in this paper is verified by repeatly experiment and then compares and analyses the measurement results.