The Study Of Trajectory Extraction And Tracking Of Steel Lines

In modern industrial production, Automated welding and cutting are being used more and more widely. Compared with the traditional manual welding and cutting, its efficiency and precision are higher. In addition, in some place which is not suitable for manual operation,for example, in some mine or some place where has radiation, Automated welding and cutting robots are being used more and more widely, at the same time, Trajectory tracking system based on vision provide a direction of development for the automated welding and cutting.First, we build the vision sensor hardware platform, and then give a description of the hardware platform (including CCD camera, camera lens, motion control card, image acquisition card) for further research. At the same time, we make an expound to the linear model and nonlinear model of the CCD camera and then give an explanation of the distortion type and the distortion model. Next, we get the internal and external parameters of the camera.Finally, according to the actual demand of the thesis, we design a camera calibration solutions based on the Two-dimensional visual trajectory tracking which is used in the industry.After we finish the hardware system, we should begin to do the image processing. This part is the most important part of the thesis. We decide to do the pre-processing operations(including linear filtering and nonlinear filtering) according to the actual of the image. Then do the gray-scale transformation and the histogram equalization processing about the image.Next, we will do the image edge detection. Article focuse its emphasis on canny edge detection. At the same time, some other edge detection operators have also been broadly described and then process the image using the variety of edge detection operator. Finally, we get the different treatment effect. For rust, we proposed the morphological methods (swell,corrosion and Opening and closing operations) to deal it and give an explanation to the effect of the image processing.For the case of the actual industrial production, outside light is always changed and we can not use the fixed threshold. So, we put forward several programs to get the threshold which is auto. Because the image processing and control movement must be done at the real-time, we present a method based on activities window (ROI). It is only to process the part of activities window which reduce the amount of computation greatly and improve the speed of the image processing. Meanwhile, because we only process the part of the activities window, some interference, for example—splash, oil stains, rust, are not need to be processed,which simplified the algorithm greatly. Finally, we use the least squares and hough transform to fit the lines.At last, we design the tracking algorithms and strategies used on the linear trajectory lines and the curve lines and give an explanation to the process of the tracking system, then the thesis shows the dynamic link of the motion control card in the lower computer and give the tracking renderings and the tracking accuracy about lines and curves under the speed of 300mm/min. Experiments show that the image processing algorithms and tracking control strategy in this thesis can meet the requirements of accuracy and speed in industrial cutting and welding.