The Chromatic Aberration 2-D Entropy Threshold Segmentation Method Based on Adaptive Step-Length Firefly Algorithm

摘要

Background: To effectively solve the segmentation problem with multi-target complex image, the chromatic aberration 2-D entropy threshold segmentation method based on Adaptive Step-Length Firefly Algorithm (ASLFA) is proposed in this paper. Methods: Firstly, the significance of image entropy value is analyzed and the threshold segmentation is proposed with maximum entropy principle. Then, in order to solve the problem of large amount and longtime of calculation in the threshold segmentation process, the improved firefly algorithm (FA) is proposed replacing the fixed step-length with adaptive step-length. Results: Finally, in order to make full use of the image information, the space distance of chromatic aberration is introduced and combined with FA. Conclusion: Contrast test of the proposed method and 2-D entropy threshold based on standard firefly algorithm (SFA) and genetic algorithm (GA) proves that the proposed method can improve the segmentation accuracy while ensuring the segmentation speed, and is suitable for fast and effective segmentation of multi-target images and complex images.