Design Of Shoe Upper Cut Pieces Matching Algorithm Based On Image Processing

Shoe upper slice sewing is an important process of automatic shoe production line.Before sewing,the cutting piece should be marked to guide sewing,so the precision of marking will directly affect the quality of sewing.At present,the marking process is usually completed by manual marking or screen printing.Manual marking accuracy is poor,efficiency is low and labor intensity is high.Screen printing takes a long time,costs a lot and is not environmentally friendly.In order to improve the production efficiency and accuracy of the marking process,this paper designs the matching marking algorithm based on image processing,and realizes the recognition and positioning of multi-objective and multi-template cutting and the calculation of the marking track.Specifically,the research content of this paper mainly includes the following aspects:1.The overall scheme of the image processing system of the marking machine is designed.,and completed the hardware equipment selection and camera calibration.Analyze the system design requirements and design the overall scheme of the system.The related hardware equipment of image acquisition system is selected.The camera imaging model and the existing camera calibration technology were studied,the camera was calibrated,the internal parameters and distortion coefficient of the camera were calculated,and the radial distortion of the camera was basically eliminated.2.An algorithm for identifying and locating vamp segments was designed,which was combined with a two-step matching algorithm of contour Hu invariant moment and feature point matching algorithm to identify and locate the segments.In this paper,the preprocessing of the cut image is firstly done to remove the interference between the press mesh and the conveyor belt joint.A matching algorithm based on contour invariant moment is used to realize fast recognition and positioning.A matching algorithm based on SURF feature points is proposed to obtain mapping relationships.Aiming at the problem that SURF algorithm has a low correct matching rate for large Angle rotating targets,SURF algorithm is improved by combining SIFT descriptor and random sampling consistency algorithm.The two-step matching algorithm neutralizes the problem of precision and speed well and realizes the fast and accurate positioning of the slice.3.The machine tool calibration algorithm was designed to optimize the streak sequence and calculate the streak data by mapping the streak trajectory.The machine tool was calibrated by drawing grid lines,and the improved Harris algorithm was used to detect corner points.The corresponding relationship between the two nozzle coordinate systems and the image coordinate systems was calculated respectively.A streaking sequence planning algorithm based on genetic algorithm is proposed to optimize the streaking track of sprinkler head by using the mutation operation of random selection exchange mutation and reverse mutation.Calculate the marking data of splines inside the slice by coordinate transformation.4.Design and develop the system software of shoe upper slice marking based on image processing.Based on the above techniques and algorithms,a matching marking software for shoe upper sections was developed by Open CVSharp and C# hybrid programming.The system can automatically identify and locate shoe upper sections,and generate the best marking path according to the optimization algorithm.The automatic marking recognition of shoe upper sections is realized,and the production efficiency and quality of the marking process are improved.