EVALUATION OF THE QUALITY OF MACHINE PARTS' MICRORELIEF BY THE METHOD OF CORRELATION-SPECTRAL ANALYSIS OF THEIR IMAGES

Abstract

The article considers a method for estimating the parameters of the microrelief of the surface of machine parts by optoelectronic and computer means, as an integral part of the technological process of manufacturing machine parts with precision surfaces. The method is based on computer processing of images of the studied microreliefs, considered as a set of realizations of a stationary random process. The number of realizations of the random process is assumed to be equal to the number of lines in the analyzed microrelief image. The microrelief image is considered as a matrix of random numbers. For this matrix, mathematical expectations, variances, standard deviations, correlation moments and the normalized autocorrelation coefficient of honey are calculated for the columns of the matrix. To conduct research on the proposed method, an optical-electronic complex was used, consisting of an instrumental microscope with a video camera and a computer for digital processing of the obtained images of the microrelief f reference samples with different roughness. The surface roughness was estimated by standard methods on a profilograph and ranged from Ra = 0.025 µm to Ra = 0.130 µm. When developing software for correlation-spectral image processing, OpenCV tools and the C++ language were used. According to the research results, it was found that the nature of the correlation functions is largely determined by the parameters of the studied microreliefs. To identify the studied microreliefs, we determined the analytical dependences of the arithmetic mean deviation of the microrelief surface profile both on the average value of the variable component of the autocorrelation function and on the values of its spectral density. It has been established that for the identification of a microrelief by optical-electronic means, the most promising is the use of the spectral density of its autocorrelation function, calculated from its halftone image. The results of applying the correlation-spectral method for assessing the microrelief the working surface of an aircraft blade are presented.