Ingot mold molten metal boundary positioning method integrating template matching and symmetric straight line segmentation

Abstract

The invention relates to an ingot mold molten metal boundary positioning method integrating template matching and symmetric straight line segmentation. Firstly, a template matching method is used fordetecting the position of one ingot casting mold, and the positions of multiple ingot casting molds are calculated according to the characteristic that the distances between adjacent ingot casting molds are equal. Then edge detection is conducted on the ingot casting area through a sobel operator by means of the color deviation between the ingot casting mold and the molten metal, and edge information inside the ingot casting is obtained. Because mirror reflection exists in the metal solution, interference shadows exist on the surface of the metal solution, a symmetrical straight line segmentation method is adopted, four edge straight lines are screened out, finally, intersection points of the four straight lines are calculated, and the rectangular position of the metal solution is obtained. Interference of factors such as shadows and illumination changes of the cast ingots can be overcome, the positions of the multiple cast ingots can be detected at the same time, and high self-adaptive adjusting capacity is achieved.

Description

technical field [0001] The invention relates to the field of computer image processing, in particular to a method for locating the metal solution boundary of an ingot mold by combining template matching and symmetrical line segmentation. Background technique [0002] In the production of non-ferrous metal ingots, the surface temperature of the non-ferrous metal solution is high, and oxide slag will form on the surface after contacting with air, resulting in the failure of the appearance quality of the metal ingot to meet the standard. At present, most metal ingot casting processes use manual slag removal, but manual slag removal is frequent and labor-intensive. After working for a long time in a high-temperature production line, there are great safety hazards. Therefore, industrial robots are used to carry out slag removal. At present, most slag removal robots use the teaching method. When the speed of the ingot production line changes or the repeat positioning accuracy of t...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the technical problem that the existing machine vision recognition technology cannot adapt to the ingot recognition on the ingot production line, the present invention provides a fusion template matching and symmetrical straight line The segmented ingot mold metal solution boundary positioning method can accurately detect the four edge straight lines of the metal solution, and has good anti-interference ability to shadows caused by light changes, good real-time performance, and strong practical significance

Images