A novel measurement method and application for grinding wheel surface topography based on shape from focus. (February 2019)

• Record Type: Journal Article
• Authors: Tang, Junjie; Qiu, Zhongjun; Li, Tianyi
• Is Part Of: Measurement. Volume 133(2019)
• Page Start: 495
• Page End: 507
• Publication Date: 2019-02
• Subjects: Grinding wheel -- Surface topography -- Shape from focus -- Sharpness function -- Three-dimensional measurement; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2018.10.006
• Languages: English
• ISSNs: 0263-2241

Abstract

Graphical abstract: Highlights: A novel measurement method for grinding wheel surface topography is proposed. A new sharpness function adapted to the grinding wheel topography is built. An improved Gaussian interpolation is built to capture the focal plane of point. The 3D reconstruction of the grinding wheel surface topography is realized. Abstract: The precision measurement of grinding wheel topography is of great significance for accurate evaluating grinding performance and improving the workpiece surface finish and grinding efficiency. In this paper, a reliable measurement method for grinding wheel topography based on shape from focus (SFF) is proposed, and in order to SSF can better adapt to the measurement of grinding wheel topography, a new sharpness function based on background subtraction and the gray level difference is constructed. The experimental results show that compared with traditional sharpness functions, the constructed sharpness function has higher unimodality (larger unimodal occupancy of total image points with 96.19%), higher contrast (smaller average occupancy of inflection point region in total image frames with 13.57%) and the shorter average operation time. On this basis, an improved Gaussian interpolation algorithm based on half-peak width is proposed to fit the sharpness evaluation curve calculated by sharpness function and capture the optimum focus position. The results show that the proposed method has better efficiency (shorter average operation time of each image points with 0.021 μs), smaller root-mean-squared-error and the sum of squares for errors, larger coefficient of determination than other curve fitting methods, which is more suitable for the proposed sharpness function to obtain the optimum focus position of image points. Finally, the three-dimensional reconstruction and measurement of grinding wheel topography are realized. The obtained results confirm the efficiency and precision of the above-mentioned method that could be a possible alternative to other methods used in grinding wheel topography measurement and then evaluating the grinding performance.