Micro scale surface texture characterization of technical structures by computer vision

Surface roughness is usually a technical prerequisite for engineering products and one of the most used significant technical index of product quality. The assessments of a mechanical part are of great importance to achieve the desired surface quality for functional performance in practice. On the other hand, the mechanism behind the formation of surface roughness is very complicated and process dependent, therefore it is very difficult to calculate its value through analytical formula simply since surface roughness is affected by many factors like feed, cutting speed and tool geometry.In this study, three workpieces were produced by conventional machining techniques. These techniques were face turning, front milling and grinding. The measurements were carried out using the confocal laser scanning type microscope. The images captured by optical measurement techniques for measuring surface roughness were analyzed by using three image processing techniques. These were line scanning, speckle and Fast Fourier Transform (FFT). Then the obtained results from images for determining roughness were compared with those obtained results from both the infinite focus microscope and the confocal laser scanning type microscope. The results from image analysis indicated that FFT analysis represented the surface roughness variation with high correlation (R2 = 0.91).

Figure optionsDownload as PowerPoint slideHighlights
► We measured three workpieces produced by face turning, front milling and grinding.
► We used confocal laser scanning microscope.
► Line scanning, speckle and FFT were used as image processing techniques.
► FFT analysis represented the roughness variation with high correlation (R2 = 0.91).