Search

Low-light image enhancement is an important task in computer vision. Deep learning-based low-light image enhancement has made significant progress. But the current methods also face the challenge of relying on a wide variety of low-light/normal-light paired images and amplifying noise while enhancing brightness. Based on existing experimental observation that most luminance information concentrates on amplitudes while noise is closely related to phases, an unsupervised low-light image enhancement method in the Fourier transform domain is proposed. In our method, the low-light image is firstly transformed into the amplitude component and phase component via Fourier transform. The luminance of low-light image is enhanced by CycleGAN in the amplitude domain, and the phase component is denoising. The cycle consistency losses both in the Fourier transform domain and spatial domain are used in training. The proposed method has been validated on publicly available test sets and shows that our method achieves superior results than other approaches in low-light image enhancement and noise suppression. [ABSTRACT FROM AUTHOR]

Very-high-resolution (VHR) bi-temporal images change detection (CD) is a basic remote sensing images (RSIs) processing task. Recently, deep convolutional neural networks (DCNNs) have shown great feature representation abilities in computer vision tasks and have achieved remarkable breakthroughs in automatic CD. However, a great majority of the existing fusion-based CD methods pay no attention to the definition of CD, so they can only detect one-way changes. Therefore, we propose a new temporal reliable change detection (TRCD) algorithm to solve this drawback of fusion-based methods. Specifically, a potential and effective algorithm is proposed for learning temporal-reliable features for CD, which is achieved by designing a novel objective function. Unlike the traditional CD objective function, we impose a regular term in the objective function, which aims to enforce the extracted features before and after exchanging sequences of bi-temporal images that are similar to each other. In addition, our backbone architecture is designed based on a high-resolution network. The captured features are semantically richer and more spatially precise, which can improve the performance for small region changes. Comprehensive experimental results on two public datasets demonstrate that the proposed method is more advanced than other state-of-the-art (SOTA) methods, and our proposed objective function shows great potential. [ABSTRACT FROM AUTHOR]

For compressive sensing magnetic resonance (CS-MR) image reconstruction, it is vital to preserve edges and textures while eliminating aliasing artifacts. The total variation (TV) regularization preserves edges well but causes staircase effect and fails to preserve textures. The high-order regularization can eliminate the staircase effect but causes edge blurring. The nonlocal TV (NLTV) regularization can preserve textures well, but causes extra artifacts and is likely to preserve the texture-like artificial structures caused by aliasing artifacts mistakenly. In this paper, we assume that the image consists of cartoon component and anisotropic component. For the cartoon component, we utilize the fractional-order TV regularization to eliminate the staircase effect and avoid edge blurring, and more importantly, we can avoid its disadvantage in texture preservation because there are no textures in the cartoon component. We utilize the NLTV regularization and the shearlet based sparsity regularization for the anisotropic component with piecewise-constant background. Without the effect of intensity inhomogeneity, the NLTV regularization can avoid preserving the texture-like artificial structures while preserving the real edges. The shearlet based sparsity regularization can provide further improvement of the image quality. Numerical experiments demonstrate that our method can eliminate the aliasing artifacts and preserve the edges and textures efficiently. [ABSTRACT FROM AUTHOR]

Abstract: The paper “staircase effect alleviation by coupling gradient fidelity term” (Zhu Lixin and Xia Deshen, 2008 []) presented a nonlinear diffusion approach using a coupling gradient fidelity term. Although such approach helps to alleviate the staircase effect to some extent, the model is not sound. Moreover the physical mechanism of this model was explained incorrectly, and there are some theoretical mistakes in their proofs and propositions. We propose a new theoretical analysis of the alleviation staircase effect using a coupling gradient fidelity term. [Copyright &y& Elsevier]