Image Enhancement Method Based On Wavelet And Adaptive Fractional Differential

An image,a frequently-used carrier for people to sense external objects,contains a large amount of valuable information.Generally speaking,it is not easy to acquire images with abundant detail features and obvious differences among weak details.With the increasing demands for multi-detail images,more and more researches,aimed at improving the qualities and weak features of images,are focused on the technologies of image detail enhancement.The methods of image enhancement will be studied from wavelet single reconstruction,adaptive fractional differential and their combination in this paper.Firstly,according to the multiresolution and time-frequency characteristic of wavelet transform,an improved image enhancement algorithm based on wavelet single reconstruction is proposed.By reconstructing the low-frequency coefficients and high-frequency coefficients of different scales after wavelet decomposition,the single reconstruction information of the same size will be obtained and then processed by the segmentation linear enhancement technique in spatial domain.After linear superposition of the low-frequency reconstruction information and high-frequency reconstruction information,low-frequency and high-frequency information can be completely separated,thereby improving the image enhancement effects of the different frequency components.Experimental results show that the algorithm is relatively flexible,which cannot only enhance the low-frequency information but the high-frequency edge information of the image.Secondly,according to the G-L definition of the fractional differential image enhancement theory,an improved adaptive fractional differential image enhancement algorithm based on image complexity is proposed.By means of the difference box dimension method in the fractal geometry,the fractal dimension,reflecting the image complexity,can be calculated.On the basis of the fractal dimension,the differentialorder of the fractional differential operation will be determined to assure the adaptive image enhancement.Experimental results show that the improved algorithm can enhance and extract the high-frequency edge information of the image to some extent while preserving the low-frequency information of the image.Furthermore,the improved algorithm can adaptively select the ideal differential order in terms of the image complexity,so that it can guarantee the optimal image enhancement effects and prevent the false edges of the image from being produced.Finally,in terms of the combination of the above two algorithms,an improved image enhancement method based on wavelet and adaptive fractional differential is proposed.According to the difference box dimension method,the fractal dimension and the differential order can be obtained.Afterwards the Tiansi template is improved,namely,the fractional filter templates of removing the horizontal direction,removing the vertical direction and removing the diagonal direction are designed.For the sake of extracting more image edge information,the wavelet coefficients of time-frequency decomposition are processed by the corresponding templates,and then the processed wavelet coefficients will be reconstructed and linearly superimposed to obtain the enhanced images and edge images.Experimental results show that this improved method can preserve the low-frequency information of the image non-linearly.The abilities to enhance and extract the high-frequency edge information are superior to those of Tiansi algorithm,other improved algorithms mentioned in this paper and the traditional edge extraction algorithms.The method can also determine the most ideal differential order adaptively,thereby achieving the optimal enhancement effects.