Plate blank cavity recognition method and system based on deep learning, medium and terminal

Abstract

The invention provides a plate blank cavity identification method and system based on deep learning, a medium and a terminal, and the method comprises the steps: collecting plate blank image data, marking the position of a plate blank in the plate blank image data, and forming an image data set; establishing a plate blank position identification network model according to the image data set, and training the plate blank position identification network model; detecting the plate blank in real time, and inputting the detected plate blank image information into the plate blank position identification network model to obtain an output result; judging whether a cavity exists in the plate blank according to the position information of the plate blank in the output result; according to the method, manual participation is not needed, working efficiency is improved, through machine learning, a deep neural network SSD-Mobilenet is used to identify the position of a plate blank in a production line. According to the method, the slab is cut, the image is cut out, the image enhancement and image binarization algorithm is adopted to judge whether the plate blank has cavities, a machine is used for replacing human eyes for recognition, the problems of low slab cavity recognition efficiency and accuracy in the prior art are solved, and the manual participation cost is reduced.

Description

technical field [0001] The present invention relates to the field of electrometallurgy, in particular to a method, system, medium and terminal for identifying slab voids based on deep learning. Background technique [0002] In the process of slab continuous casting and steel rolling in the field of iron and steel smelting, it is necessary to inspect the slab after the sintering process to determine whether there are cavities in it. Internal defects, once there are cavities in the slab, will fluctuate the quality of the final slab and affect the use value of the slab, so the identification of cavities in the slab is particularly important. [0003] However, at present, the judgment of cavities in slabs is usually still judged by manual observation, which not only has low work efficiency, but also cannot guarantee the accuracy of recognition results, which will affect the quality of slabs in the steelmaking process. Therefore, a new method for identifying slab voids is needed...

AI Technical Summary

Problems solved by technology

[0003] However, at present, the judgment of cavities in slabs is usually still judged by manual observation, which not only has low work efficiency, but also cannot guarantee the accuracy of the recognition results, which will affect the quality of slabs in the steelmaking process

Images