Research on Denoising Localization Algorithm for Positioning Laser Spots and Noise Spots Interlaced Image

In the industrial production, both ceiling lighting and sunlight through the ceiling in the workshop can cause noise spots in the image acquired in the CCD laser online positioning system. Moreover, these ambient noise spots are interlaced with the positioning laser spots in the image, which makes it difficult to obtain the position information of the positioning laser spots accurately. Aiming at this problem, in this paper, according to the geometric characteristics of both positioning laser spots and ambient noise spot, a denoising localization algorithm for positioning laser spots and noise spots interlaced image is proposed, and the cylindrical bar positioning system is used as the experimental object to verify the algorithm. The algorithm steps are as follows, firstly the noise spot is located according to the geometric characteristics of the two kinds of spots, and the non-overlapping region of the two kinds of spots are identified at the same time. Secondly, the noise spots in the non-overlapping area are eliminated. Finally, the existing Blob positioning algorithm is slightly modified to complete the accurate extraction of the position coordinates of the positioning laser spots pixel. The algorithm can effectively eliminate the ambient noise spots on the premise of ensuring that the positioning laser spots are not damaged, and accurately realize the measurement of the position coordinates of the positioning laser spots pixel.