Research On Wear State Detection Of Ring-throwing Machine Disk Based On Image Method

The wear condition of asphalt disc of large ring polisher has a great influence on polishing time of optical glass elements.In ultra-precision polishing,asphalt disk is used as a medium to carry polishing fluid and directly participate in the polishing process.The change of its surface condition will directly affect the polishing time and polishing quality.Therefore,the wear state detection of asphalt disc of ring polisher can grasp the situation of asphalt disc and take corresponding measures to optimize the polishing process and speed up the polishing process of optical glass.Aiming at the difficulty of asphalt disc wear detection,considering the actual polishing process,image texture analysis method is used to realize the convenient non-contact detection of asphalt disc wear state of ring polisher.The main contents of this paper are as follows:1)According to the situation of the asphalt disk image collected in this experiment,image preprocessing methods such as gray scale,histogram equalization and median filtering are used to preprocess the asphalt disk image.2)The region of interest is segmented from the captured image by using binary image operation rules in the way of "seeking maximum connected region".Based on Retinex theory and gamma function,the illumination component of the image is extracted and the phenomenon of uneven illumination is removed to eliminate the influence of uneven illumination on the quality of the disk image,so as to ensure the accuracy of image texture analysis.3)Based on the gray level co-occurrence matrix method,the texture feature parameters of the pre-processed asphalt disk image are extracted and analyzed.Firstly,according to the characteristics of asphalt surface image,three construction parameters of gray level co-occurrence matrix are determined: gray level,step size and direction of generation.Then,four characteristic parameters of gray level co-occurrence matrix in asphalt disk image are extracted: energy,entropy,contrast and correlation,and the four characteristic parameters are analyzed.The relationship between the characteristic parameters of gray level co-occurrence matrix and the wear state of asphalt disks is established.4)Using regression analysis method,a detection model of asphalt disc wear stateis constructed,and the mathematical expression of the relationship between asphalt disc texture characteristic parameters and asphalt disc wear state detection is fitted.The feasibility of this method is verified by many experiments.