Adaptive Clipping Histogram Segmentation Equalization Image Enhancement

Image enhancement is an image processing technology that improves the visual effect and clarifies useful information of images.It has a significant impact in the field of pattern recognition and computer vision.Among many image enhancement algorithms,image enhancement algorithms based on histogram equalization have been widely concerned by many researchers.The algorithm can expand the grayscale distribution range of the histogram according to the conversion function to achieve the purpose of enhancing the image.However,the conversion function itself has certain limitations,so it is easy to cause problems such as image contrast excessive enhancement and average brightness shift.At the same time,only the intensity information of the pixels is used as a sample to construct the histogram,but the spatial position information of the pixels is not considered.In order to overcome the above problems,this article proposes three improved algorithms,as follows:1.Aiming at the problem that the histogram clipping value and histogram segmentation threshold in the HE image enhancement algorithm cannot achieve adaptive selection,this paper proposes an adaptive contrast-limited sub-histogram equalization(ACLSHE)based on S-type fuzzy membership function for image enhancement.The algorithm uses an S-type membership function to calculate the clipping value of a histogram,and the algorithm realizes the adaptive selection of the histogram clipping value;And the algorithm uses the global peak search algorithm to adaptively find a histogram segmentation threshold.Experimental results show that the ACLSHE algorithm has better enhancement results for natural images,MRI brain images,and near-infrared images,and the algorithm improves the image quality evaluation index value.2.In view of the problem that most image enhancement algorithms only consider pixel intensity characteristics and ignore their spatial information,the intuitionistic fuzzy set theory is used to describe the position information,ambiguity and unknown information between image pixels,this paper proposes adaptive intuitionistic fuzzy dissimilar histogram clipping(AIFDHC)image enhancement algorithm.The algorithm uses "voting model" theory to extend the fuzzy dissimilarity histogram to the intuitionistic fuzzy dissimilarity histogram.Image fuzziness and uncertainty are introduced into the image enhancement process.At the same time,the algorithm uses hesitation to correct the detailed image obtained by the guided filtering and enhance the detailed information in the image.The experimental results show that the AIFDHC algorithm can obtain a more natural image visual effect after introducing the image pixel spatial information,and the image quality evaluation index has been significantly improved.3.Compared with the triangular fuzzy membership function,the Gaussian fuzzy membership function is more in line with the nonlinear characteristics of an image histogram.At the same time,in order to further enhance the detailed texture characteristics of an image,this paper proposes the intuitionistic fuzzy histogram detail enhancement(IFHDAE)algorithm.The algorithm uses Gaussian fuzzy membership function and "voting model" theory to construct an intuitionistic fuzzy histogram.Then the standard deviation is used to realize adaptive enhancement of details.Experimental results show that the IFHDAE algorithm can not only make full use of pixel spatial information,but also maintain better detail information in the enhanced image.