Research On Surface Defects Detection Of Piezoelectric Ceramic Chips

Piezoelectric ceramic chip is a kind of engineering ceramic material,It is the core component of various sensors,transducers and electronic devices because of its piezoelectric property and good mechanical and electrical energy characteristics,and is widely used in various fields.In the process of manufacturing piezoelectric ceramic chips,due to the impurities in raw materials and improper processing technology,defects such as spots,cavities and cracks are easy to occur on the surface,resulting in degradation of material properties and serious impact on product quality.Manual visual inspection is the main method adopted by domestic production enterprises,the etection efficiency is low,accuracy is poor,missed detection,false detection often occurs.Therefore,in order to improve the quality of piezoelectric ceramic products,improve the detection efficiency and accuracy,it is of great significance to study an automatic detection system.In recent years,with the continuous maturity of computer technology and sensor technology,machine vision detection method provides a new way for defect detection.Based on the study of the material and physical properties of piezoelectric ceramic chip,the causes and visual characteristics of the spots and cavities,the overall design of defect detection is determined.Through theoretical calculation and imaging analysis,the model selection of light source,camera and lens and the design of space position in the imaging system are completed.Combined with the motion control platform,the defect detection hardware system is built.In view of the irrelevant information in the collected image,some preprocessing algorithms,such as image denoising,threshold segmentation,area feature selection,are used to segment and extract the piezoelectric ceramic area.For the spot defect,the dynamic threshold and area feature are used to extract the defect area.Aiming at void defects,the method of Fourier transform frequency domain analysis is used to enhance the contrast of defects,and neighborhood enhancement and area selection are used to extract void defects in space domain.Based on the hardware model of this project,the camera driver acquisition program,motion control program and interface display program are written in Visual Studio 2012 integrated development environment.The multithreading technology is used to solve the problems of interface response,image acquisition,motion control and defect detection running at the same time.Through the test of the self built piezoelectric ceramic chip library,the test results are statistically analyzed.The experimental results show that the accuracy of the system is 95.16% and 93.18%,respectively.The detection effect is good and has a certain application value.