Search

Low-light images captured at night often suffer from improper exposure, color distortion, and noise, which degrades the image quality and have a negative influence on subsequent applications. Many existing deep learning-based methods enhance low-light images through spatial domain, which may sacrifice the original image information. In this paper, we put forward a deep learning network for enhancing low-light images based on wavelet transform. We utilize the wavelet transform to divide the image into various frequency scales and then analyze the frequency characteristics of different low-light images in the wavelet domain. The proposed network comprises a low-frequency restoration subnet and high-frequency reconstruction subnet that uses an optimal coefficient of wavelet decomposition to construct a frequency pyramid. Furthermore, we utilized different attention mechanisms to extract frequency information from different images, gradually restoring the brightness information and details of low-light images. Additionally, we utilized a self-constructed multi-scale exposure low-light image dataset for training. Numerous experiments on publicly accessible datasets and our established dataset show that the proposed approach quantitatively and qualitatively surpasses state-of-the-art approaches, particularly for real and complex low-light scenarios. Furthermore, our method produces better visual effects than others and performs well in real-time and real-word downstream vision tasks.

Abstract Geo-materials may present varying mechanical properties under different stress paths, especially for tunnel excavation, which is typically characterized by the decreased radial stress and increased axial stress during the complex loading and unloading process. This study carried out a comparative analysis between the loading and unloading model testing, which was then combined with PFC2D simulation, aiming to reveal the fracture propagation pattern, microscopic stress and force chain distribution of the rock mass surrounding the tunnel. Comparisons of extents and development of tensile strain between loading and unloading testing results were made. The overall stability, the integrity of rock mass, and the failure pattern transition under loading and unloading processes were systematically examined. In addition, for the two unloading cases with different vertical stresses imposed, the failure patterns were both identified as the collapse of the V − shaped extruded sidewall, due to the coupling of the shear failure and the vertical tensile failure in the sidewall wedge.