Font Size: a A A

The Research On Engine Cylinder Sorting And Identification System Based On Machine Vision

Posted on:2020-08-05Degree:MasterType:Thesis
Country:ChinaCandidate:H ZhouFull Text:PDF
GTID:2428330575980467Subject:(degree of mechanical engineering)
Abstract/Summary:PDF Full Text Request
Machine vision technology is a favorable weapon for the country to achieve Industry 4.0 goals.This paper combines traditional engine block sorting with machine vision technology to add the intelligent "eyes" and "brains" of machine vision to traditional industries.Adding machine vision not only improves the efficiency of engine block sorting,but also improves accuracy,and has the advantage of being irreplaceable in terms of sustainability and cost reduction.This paper designed the overall structure and system scheme based on machine vision recognition sorting system,and built a visual system platform.Software and hardware selection for the solution involved in the vision system.The entire camera system is calibrated to establish a theoretical connection between the pixel coordinate system,the image coordinate system,the camera coordinate system,and the world coordinate system.Explain how to perform camera calibration under HALCON and what operator to use for calibration.The relationship between the pixel coordinates on the image and the actual world coordinates.Templates are created by selecting a specific pose of each type of engine block as a template.The photos are collected in real time,and the image is smoothed by mean filtering to enhance the image contrast and make the contour easier to extract.The ROI area is segmented,and the effective area is extracted for template matching.Corresponding engine block categories and relative rotational angles and center positions are derived via template matching.Perform OCR(Optical Character Recognition)to perform median filtering on the captured image,enhance image contrast,grayscale threshold,and perform corrosion image morphology processing.The selected OCR characters are sorted and compared with the OCR character template.The image processing is divided into two-step screening in order to sort the engine blocks of the same type and the same number and then stack them.The whole circuit system is designed to determine the secondary distributed structure of the upper computer and the lower computer.The PC is determined as the upper computer,and the motion control card and the image acquisition card are respectively used as the lower part of the electronic control part and the image processing part.And hardware selection of the motor driver.With VC++6.0 as the development platform,combined with the interface function of MFC.The configuration of HALCON under the whole VC and the method of loading the motion control card into the VC are studied.The entire system is integrated into the MFC interface display,and the software programming is completed,so that the entire software has control functions and real-time monitoring functions.The software experiment results show that all functional modules can work normally,the performance indicators meet the requirements of the engine cylinder sorting and identification system,and the robot realizes intelligent sorting with the participation of visual software.
Keywords/Search Tags:Machine vision, engine block, sorting recognition, template matching, optical character recognition
PDF Full Text Request
Related items