Infrared image segmentation based on 2-D maximum fuzzy entropy and intelligent genetic algorithm

Abstract

AbsimctThis paper presents a fast and effective segmentation. method for Forward Looking Infra-Red (FLIR) 1magei-based:on fuzzy filtering, the principle of 2-D maximum fuzzy entropy. and. intelligent genetic algorithm. A new fuzzy operator is applied Wthe enhancement of noisy images. Basing on fuzzy relation and maximum l u n y entropy principle, we perform fnzzy partition on a 2-D histogram. A new intelligent genetic algorithm (IGA), which applies an intelligent crossover (IC) based on orthogonal arrays (OAS), Is proposed to search the .optimal wmbination of the fumy parsmetem finally used to segment the image. Experiment results show that the proposed method can depress the Gaussian (-like) noise effectively, segment the infrared image automatically and properly, and i s faster than the conventional genetic algorithm and exhaustive search.