Forward-and-backward diffusion processes for adaptive image enhancement and denoising

Abstract

Signal and image enhancement is considered in the context of a new type of diffusion process that simultaneously enhances, sharpens, and denoises images. The nonlinear diffusion coefficient is locally adjusted according to image features such as edges, textures, and moments. As such, it can switch the diffusion process from a forward to a backward (inverse) mode according to a given set of criteria. This results in a forward-and-backward (FAB) adaptive diffusion process that enhances features while locally denoising smoother segments of the signal or image. The proposed method, using the FAB process, is applied in a super-resolution scheme. The FAB method is further generalized for color processing via the Beltrami flow, by adaptively modifying the structure tensor that controls the nonlinear diffusion process. The proposed structure tensor is neither positive definite nor negative, and switches between these states according to image features. This results in a forward-and-backward diffusion flow where different regions of the image are either forward or backward diffused according to the local geometry within a neighborhood.

Color Image Contrast Enhancement Based on Spatial Transformation and Expert Fuzzy Technique

“…The second is to highlight and reconstruct the image details [5]. The present methods of image enhancement technology are various, such as: grayscale transformation processing method [6];Spatial filtering measures [7];The method of frequency domain image enhancement and multi-scale analysis based on wavelet transform [8];Multi-image fusion enhancement technology for multi-sensor or single sensor [9];Image defog enhancement algorithm [10];Image denoising and enhancement based on adaptive diffusion equation [11];Image low contrast enhancement based on RETINEX model [12];The new adaptive cuckoo search algorithm is used for satellite image contrast enhancement [13] and other technologies. Most image enhancement methods are based on experience, and can be satisfied by repeated modification of the parameters.…”

Edge and contrast preserving in total variation image denoising

“…Guy Gilboa et al [23] proposed forward and backward diffusion (FBD) to simultaneously remove the noise and enhance the contrast.…”

Image Denoising Using Sparse Representations

“…A very common assumption is that the present noise is additive zero-mean white Gaussian with standard deviation σ. Many solutions have been proposed for this problem based on different ideas, such as spatial adaptive filters, diffusion enhancement [1], statistical modeling [2], transfer domain methods [3], [4], order statistics [5] and yet many more. Among these methods, methods based on with sparse and redundant representations has recently attracted lots of attentions [8].…”