Design Of Visual Inspection System For High Temperature Wire Automatic Production Line

Engine wiring harness as one of the indispensable parts of the automotive engine assembly,plays a key role in connecting the circuit and transmitting signals,and the quality of the wiring harness affects the assembly of subsequent working circuits and even the stability of subsequent engine work.At present,the domestic manual production efficiency is low,with large differences and poor stability,and the quality inspection also relies on manual,with low accuracy,unreliable quality and high cost,which cannot meet the needs of mass production and inspection of wiring harnesses in high efficiency and real time.In this paper,we take engine wiring harness as the research object and carry out the research and design of automatic production line design and visual inspection system for high temperature wire.According to the engine wire harness production requirements,the production process is divided into 10 production steps,and the functions of feeding,testing and control of each production step are completed.The actual production is carried out after debugging the production line,which proves the rationality of the production line design.The quality inspection of the whole production line,including the wire head,terminal crimping and sleeve 3 parts.Among them,the distance between the sleeve and the terminal is prone to quality problems,using machine vision to detect the distance between the sleeve and the terminal,this paper takes the detection system as the representative,the overall system scheme design,hardware equipment selection,in the industrial environment to complete the detection system hardware construction.Design and development of the detection system software interface,written in Visual Studio 2019 integrated environment,using C# language to independently complete the software underlying framework.In the image processing part,call the Open CV function library to write the image processing algorithm related to visual dimension measurement.For image edge detection and localization,the Zernike moment subpixel edge localization algorithm is introduced,and the Zernike moment subpixel edge localization algorithm is improved to enhance the localization accuracy.Using the principle of fitting straight lines by Haff transform,the edges of the sleeve and terminal in the image are fitted with straight lines for distance measurement.The camera calibration technique is studied,the camera calibration principle is derived and with the help of Matlab software,the industrial camera is calibrated using the Zhang Zhengyou calibration method,the camera parameters are obtained,and the conversion of pixel values in the image to the actual size measurement is realized.The actual production line was debugged and the whole production line was able to carry out automated production of 20-24 wire harnesses per minute,which is more than 5 times more efficient than manual production.The standard deviation of the measurement results after the algorithm improvement was reduced by 80%,and the detection error could be controlled within 0.02 mm,which illustrates the accuracy and effectiveness of the improved Zernike moment sub-pixel edge positioning algorithm proposed in this paper,which meets the detection requirements and improves the productivity.