Automatic identification and grading method of center segregation defect in low magnification structure of continuous casting slab

Abstract

The present invention provides an automatic identification and rating method for low-magnification microstructure center segregation defects of continuous casting slabs, including: S1: preprocessing the grayscale image of the low-magnification microstructure of continuous casting slabs, including distortion correction, cutting white background and filtering processing ; S2: Aiming at the central area where the horizontal area of ​​the grayscale image overlaps with the vertical area, the central area image is obtained by segmenting the grayscale image; S3: Detecting and marking the connectivity in the binary image corresponding to the central area image The area separates the suspected central segregation area from the crack area in the connected area; S4: extracts the features of the suspected central segregation area, identifies the central segregation area and removes the interference area; S5: trains the BP neural network classifier model to identify the central segregation area and rating. The technical scheme of the invention solves the problem that the traditional digital image processing method is not suitable for the detection of the continuous casting slab defects which are redundant and easy to intersect with surface defects.

Description

technical field [0001] The invention relates to the technical field of continuous casting slab defect detection, in particular to an automatic identification and rating method for continuous casting slab low-magnification structure center segregation defects. Background technique [0002] In the process of steel production, it is necessary to track and monitor the internal quality information of the produced slab. The detection method of the internal quality of the slab is to observe the low-magnification sample after the acid etching of the continuous casting slab to determine the type and level of surface defects. At present, the method of inspection and rating of low-magnification defects of cast slabs still adopts the manual rating mode, that is, the pictures of low-magnification samples of cast slabs are compared with the national standard YB / T 4002-2013 "Continuous Casting Billet Low-magnification Structural Defects Grading Chart" for visual inspection or manual measur...

AI Technical Summary

Problems solved by technology

At present, the detection and rating of central segregation of low-magnification defects in continuous casting slabs generally adopts visual inspection, because there are multiple textures on the continuous casting slab, and the area of ​​central segregation itself is not large compared with the continuous casting slab surface At the same time, there are also factors such as cracks, oil stains, shrinkage cavities and other interference

Images