Automatic Recognition and Calculation of Exposed Area of Molten Steel in Ladle Based on Image Processing

In view of the oxidation of exposed molten steel in the ladle and its influence on the quality of molten steel, au automatic recognition and calculation method of exposed area of molten steel in ladle based on image processing was proposed. The sample set of ladle liquid surface image was constructed, and the gray processing, bilateral filtering, adaptive histogram equalization and threshold processing were carried out on it to obtain the binary image of molten steel and no molten steel, and then the contour extraction and area calculation were carried out. The effects of different image denoising, image enhancement and threshold processing methods were compared through simulation experiments. The results show that the image processing method of integrating bilateral filtering, adaptive histogram equalization and global threshold processing can quickly and accurately identify and calculate the exposed area of molten steel. The bilateral filtering method can better protect the information of image edge and structure. The image processed by the adaptive histogram equalization method contains more image information, which can better fuse the image quality, the image similarity is higher, the distortion degree is lower.