Laser stripe center extraction method base on Hessian matrix improved by stripe width precise calculation

Abstract

The line laser measurement technology based on triangulation is widely used in the field of complex surface topography measurement. The key technology of line laser measurement is laser stripe center extraction. For high reflective surfaces, due to the low surface roughness, some areas of the laser stripe image are overexposed, resulting in uneven stripe width distribution, asymmetric light intensity distribution of the stripe cross section, and difficulty in extracting the center of the stripe. In this paper, Laser stripe center extraction method based on Hessian matrix method improved by stripe width is proposed. Firstly, the laser stripe image is preprocessed, the influence of noise in the image is alleviated by image filtering and adaptive threshold segmentation, and the region of interest of the image is extracted to reduce the amount of subsequent calculation. The laser stripe width is calculated based on BPNN and divided into different width intervals. The Gaussian kernel is calculated according to the light stripe width and the Hessian matrix is constructed. The Hessian matrix method is used to roughly extract the center coordinates of the light stripe. Based on the continuity of the light stripe, the center coordinates of the light stripe are weighted and corrected to realize the smooth optimization of the center coordinates of the light stripe. The experimental results show that the proposed light stripe center extraction method is superior to the traditional method under different surface roughness and exposure time conditions, which can effectively improve the measurement accuracy of the sensor.