The Development Of Sole Finishing Contour Detection System Based On Machine Vision

At present,most shoe-making enterprises use processing equipment with low efficiency and poor processing quality.Finishing the edge of the shoe sole is a key process in the manufacturing process and directly affects th e quality of the finished product.Due to the rapid development of machine vision technology,machine vision provides effective technical support for automated manufacturing and production.After analysis of the above processing problems,according to the technical requirements of sole contouring,this paper developed a machine vision-based sole contouring detection system,combined with machine vision and numerical control technology,to study the outline extraction of finished shoe soles and CNC machining tools.The trajectory generation method solves the technical difficulties in finishing automation of shoe sole edges.According to the process requirements and the user to develop the overall requirements of the vision system,from the perspective of econ omical efficiency and feasibility,in the selection of the visual hardware,according to the speed and accuracy requirements calculations to select the industrial camera,lens and light source that meet the production requirements.From the actual processing requirements,a reasonable arrangement of the detection system in the production line installation.In the image preprocessing,different methods of smooth filtering and threshold segmentation are compared.Through Gaussian filter processing,the noise in the image is effectively removed,and the image edge smoothing effect is good.The largest inter-class variance threshold method solves the segmentation problem of the image foreground and background well.In order to accurately extract the edge informa tion of the sole,all the contours in the image are searched,and the remaining interference edges are eliminated according to the basic geometrical characteristics of the contour so as to accurately retain the information of the contour of the sole.Because the strut mechanism inevitably appeared in the image,Shi-Tomasi was used to find the corner point algorithm to screen out the intersection point between the rough sole and the strut and remove the interference edge.Completing the notch with a straight line connection does not fully satisfy the process requirements.In order to enable the smooth transition of the sole edge,this paper proposes a method of least squares fitting the quadratic curve to complete the contour repair.According to the requirements of the finishing process,this paper proposes three methods for generating tool paths: mathematical modeling,image morphology,and outline point circle.The advantages and disadvantages of the three methods are compared.Finally,the tool is selected using the method of contour point circle to produce the tool.Processing path.According to the processing requirements of the numerical control system,the coordinates of each point of the tool processing path are converted into the form of polar coordinates in the form of rectangular coordinates,and all coordinate data of the points are sent to the numerical control system through the serial communication.Finally,based on the above theoretical research,the software of sole contouring detection system based on machine vision was developed.Through the experimental data and the actual production analysis,the tool path is calculated and optimized in sections,the shortage of system software is improved,and good expected results are obtained.