Research on ICT Image Segmentation Based on Firefly Algorithm

ICT is a commonly used non-destructive testing method nowadays, but due to the specificity of mechanical tissue materials, reconstructed images are prone to metal artifacts. This requires precise segmentation of high-density metal areas in ICT images to reduce metal artifacts in reconstructed images. Traditional image segmentation methods have the problems of high computational complexity and low segmentation accuracy when processing ICT images. Although the firefly algorithm in swarm intelligence algorithms has good optimization and search capabilities, it faces challenges such as difficulty in parameter selection and premature convergence in practical applications. This article uses an improved firefly algorithm to preprocess ICT projection data, successfully segmenting high-density metal region images, namely regions of interest, and further studying the problem of metal artifact correction in ICT reconstruction, providing new ideas and methods for improving ICT image quality and detection accuracy.