Corrosion Damage Assessment of AerMet100 Steel Based on Image Analysis

Abstract

This paper aims to evaluate the corrosion damage of AerMet100 steel based on image analysis technology. Through the accelerated corrosion test of AerMet100 steel, the corrosion morphology under different corrosion ages was obtained. Based on MATLAB software, median image filtering, grayscale transformation and fuzzy enhancement are used to preprocess the corrosion image. Then the fractal theory was used to determine the optimal grayscale threshold for the corrosion image of AerMet100 steel. Then the binary image processing technology was used to extract the features and calculate the corresponding degree of corrosion damage. In the accelerated corrosion of 3a and 6a, the corrosion morphology of AerMet100 steel was pitting corrosion, and the corrosion damage was lighter; thereafter, it gradually developed into uniform corrosion, and the corrosion damage was aggravated. For the specimens with different corrosion ages, the fractal theory can be used to obtain the corresponding optimal grayscale threshold and the corresponding fractal dimension. In the initial stage of corrosion, the corrosion damage degree of the specimen increases rapidly with the increase of the corrosion time. In the later period of corrosion, the increase of the corrosion damage tends to be gentle. Based on image analysis technology, it can provide reference for corrosion damage characterization and evaluation of AerMet100 steel.