A method and system for identifying a textured metal solidification region

Abstract

The invention discloses a textured metal solidification area identification method, which is characterized by including the following steps: Step 100, correcting and segmenting the metal low-magnification sample image; Step 200, using local automatic method based on grayscale mean square error The adaptive threshold binarization algorithm extracts the binarized features of the texture in the sample image; Step 300: Use a lattice direction measurement algorithm to describe the direction features of the texture and use a horizontality-based directional filtering algorithm to orient the texture features. Filtering; Step 400: Complete the division of the solidification area according to the texture direction information. The present invention uses digital image processing technology to perform preprocessing, feature extraction, pattern recognition and other operations on the low-magnification structural solidification area image of the casting slab to achieve automatic identification and quantification, improve detection accuracy, while reducing manual labor intensity and improving production efficiency.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a method and system for recognizing textured metal solidification regions. Background technique [0002] The quality of the internal quality of the continuous casting billet directly affects the quality and performance of subsequent products. In the production process, various factors (such as non-metallic impurities, gases in molten steel, improper temperature control, process defects or operator errors, etc.) will cause low-magnification defects inside the continuous casting slab and affect the solidification structure of the continuous casting slab. Structure. The structure of the solidification structure of the continuous casting slab mainly reflects the morphology of the crystallization of the slab, and the difference in crystallization conditions will cause corresponding changes in the structure of the solidification structure. By identifying the defects and reg...

AI Technical Summary

Problems solved by technology

The current macro-inspection method is to manually divide and analyze the condensation area of ​​the low-magnification structure of the slab based on personal experience. The manual inspection method has low reliability, large errors, lagging results, lack of database support, and each quality inspector. The level of experience is uneven, and the judging criteria are different, so subjective factors will affect the test results

Images