Boundary localization method of ingot metal combining template matching and symmetric straight line segmentation

To solve the problem of multi-objective localization of non-ferrous metal ingots and the shadow interference caused by specular reflection on the boundary localization of metal solution, a boundary detection method combining template matching and symmetric linear segmentation is proposed in this paper. Firstly, the position of a single ingot tool is detected by the template matching method, and the position of multiple ingot tools is calculated according to the characteristic of equal distance between adjacent ingots. Secondly, the Sobel operator is used to detect the edge of the ingot region. In view of the presence of shadow interference on the surface of the metal solution, a symmetric linear segmentation method is proposed to extract the four edge lines of the metal solution. Finally, the intersection of the lines is calculated to determine the boundary of the metal solution. The experimental results demonstrate that the method can overcome the interference of factors such as shadows on the solution surface and changes in illumination. The method also can detect multiple ingot positions simultaneously, with strong adaptive adjustment capability.