Research On Detector Base Detection System Based On Machine Vision

China's relay production in 2013 accounted for 50 percent of the world.In recent years,China's relay industry from the product quickly turned to quality.However,for the domestic manufacturing of the relay base,the traditional manual quality testing methods cannot be online detection and accuracy is not high.The efficiency is far from meeting the requirements of the market.Therefore it is necessary to develop a high-precision and high-efficiency on-line inspection system.In this paper,an in-depth study is done based on the visual inspection technology.The main research work is as follows:1.According to the requirements of the relay base defect identification and size detection,a whole scheme of combining the defect identification and the size measurement system with a plurality of cameras is proposed,which can realize the automation of the on-line detection system and Integrated operation.2.Using the normalized correlation coefficient similarity measure method to make it better adapt to the linear change of illumination.Adopt the image pyramid hierarchical search strategy on the search strategy to improve the search speed.Create multi-directional template to improve the accuracy of rotation or zoom image matching.The improved template matching algorithm based on gray value solves the problem that the existing defect recognition algorithm is affected by the external environment,the matching effect is poor and the detection speed is slow.3.Use pixel-level edge detection to get "rough" contour for the target image calibration and threshold segmentation processing.And then select a rectangular area by the contour point as the center to determine the region of the gray value of the pixel gradient position Coordinate and grayscale.Select the cubic spline function to interpolate and finally calculate the sub-pixel coordinates.The improved sub-pixel edge detection algorithm is proposed to solve the problems that traditional detection method can not be quantified,low precision defects,the measurement accuracy and speed.4.Based on the Windows system Visual Studio 2015 platform development and design software system framework including the camera control module,create a template module,defect identification module,size measurement module and automatic operation module.And the entire system hardware and software were tested.Mainly for the matching accuracy and matching speed and measurement accuracy and detection time were evaluated and verified.The research and development of the machine vision based relay base detection system can realize the accurate identification of the defect and the high precision measurement of the size.And it has high market application value.