Image Thresholding Segmentation Method Based on Adaptive Granulation and Reciprocal Rough Entropy

Xiying Chen,Caihui Liu,Dehua Xie,Duoqian Miao

DOI: https://doi.org/10.1016/j.ins.2024.121737

IF: 8.1

2024-01-01

Information Sciences

Abstract:The rough set-based rough entropy image thresholding algorithm can effectively handle the uncertainty information in the image. This method has two important steps: the design of the granulation scheme and the selection of the threshold value. However, existing methods do not incorporate information from the image to achieve adaptive granulation and are not precise enough to represent image boundaries. Therefore, a novel adaptive granulation rough entropy image thresholding method is proposed. First, a grayscale value commonality index (GVCI) is introduced by integrating grayscale information of the image, and a histogram-based adaptive granulation method is proposed. Second, a variable precision rough set-based reciprocal rough entropy threshold selection method is proposed. The variable precision rough sets (VPRS) enable a more precise representation of the boundary between the target set and background set of the image, and the optimal threshold is obtained by minimizing the reciprocal rough entropy. Finally, a variable precision rough set-based adaptive granulation reciprocal rough entropy image thresholding method is proposed. In the experiments, the proposed algorithm is compared with six existing image segmentation algorithms, and its effectiveness is verified through qualitative and quantitative analysis.