Abstract:Aiming at the shortcomings of the classic histogram equalization image enhancement algorithm, such as excessive contrast, uneven brightness distribution and low detail information, an adaptive intuitionistic fuzzy dissimilar histogram clipping image enhancement algorithm is proposed. Firstly, based on the ``voting model'' of intuitionistic fuzzy sets, this paper defines the concept of an intuitionistic fuzzy dissimilar histogram to extract the spatial location information of image pixels. Secondly, an S-type membership function is used to adaptively clip the intuitionistic fuzzy dissimilar histogram. Then, a segmentation strategy is used to equalize the clipped intuitionistic fuzzy dissimilar histogram. Finally, the hesitancy of the intuitionistic fuzzy set is used to characterize the unknown information of the original image, and modify the detailed image obtained by the guide filtering, thereby the rich detailed information of an image is retained. The experimental results for three types of images, namely natural images, MRI brain images and near-infrared images, show that the proposed algorithm efficiently enhances contrast, retains the details, yields natural visual quality images, and improves an image quality evaluation index.