Research On Defect Detection Method Of Blind Hole Machining Based On Image Processing

With the rapid development of industrialization,defect detection has become an important step to ensure product quality.However,in the process of manufacturing blind holes,due to the influence of production technology and human error,the produced blind holes will appear scratches,cracks,elliptic holes and other defects.In this paper,on the basis of looking up the defect detection technology at home and abroad in detail,a blind hole machining defect detection method based on image processing is proposed.First of all,after comprehensive investigation and analysis of machine vision detection methods at home and abroad,the defect detection method based on image processing is selected as the research method of this paper.Firstly,the selection of industrial camera,optical lens and other hardware equipment as well as the lighting mode of light source is determined,and an image acquisition device is established.CCD industrial camera is used to collect the blind hole image,and the image is transmitted to the computer for image processing.Secondly,in order to meet the requirements of blind hole defect detection,the algorithm design of blind hole machining defect detection system is carried out.Image processing algorithm is the core part of the whole defect detection,this paper focuses on the image enhancement algorithm,image segmentation and edge detection algorithms,and the image preprocessing process experiments.Histogram equalization is used to enhance the contrast of the original image.In order to eliminate the effect of noise on the image,smooth filtering is used to denoise.This paper introduces the traditional threshold segmentation algorithm and the basic operation of morphology,and uses the threshold segmentation algorithm to carry out experiments on the blind hole defect image.In order to suppress the noise and recognition defects caused by image binarization,the connected domain labeling method is selected.Then an improved algorithm based on Canny edge detection is proposed and the experimental results are compared and analyzed.After studying the feature extraction and classification of defects,support vector machine is selected to classify defects,and finally achieve the purpose of complete classification of defects.Finally,this paper uses Open CV software to program the blind hole defect detection,tests the blind hole defect image with the above method,analyzes the experimental results,and further optimizes the algorithm.Detection of blind hole machining defects based on image processing technology can not only improve production efficiency and ensure production quality,but also reduce the working time of workers and achieve the purpose of improving work efficiency.