Abstract:We present a streamlined and effective fusion-based approach for enhancing images with low illumination, utilizing a combination of well-established image processing approaches. Initially, we utilize an illumination estimation algorithm relying on morphological closing to decompose a given image into a reflectance image as well as an illumination image. Subsequently, we define two inputs indicating luminance improvement as well as contrast enhancement, achieved by applying the sigmoid function as well as adaptive histogram equalization to the initially decomposed illumination. By designing two weights derived from these inputs, we generate an adjusted illumination through multi-scale fusion, combining the derived inputs with their respective weights. By employing an appropriate weighting as well as fusion strategy, we integrate the benefits of various methods to generate the adjusted illumination. The ultimate enhanced image is acquired by compensating for the adjusted illumination back into the reflectance. Through this synthesis, the improved image embodies a balance between enhancing details, improving local contrast, and preserving the natural appearance of the image. In the suggested fusion-based framework, images captured under various conditions of weak illumination, such as backlighting, non-uniform illumination, as well as nighttime scenarios, can be improved. Extensive experimentation has demonstrated that the proposed approach delivers comparable or superior performance when compared to state-of-the-art competing techniques, as evidenced by both qualitative as well as quantitative evaluations. Furthermore, this includes a comprehensive overview of different quality evaluation approaches for enhanced images, along with discussions comparing various algorithms. Lastly, the present research progress is summarized, with potential future research directions are proposed.

Weighted Sparse Representation and Gradient Domain Guided Filter Pyramid Image Fusion Based on Low-Light-Level Dual-Channel Camera

Detail Preserving Exposure Fusion for A Dual Sensor Camera

Fusion of Low-Illuminance Visible and Near-Infrared Images Based on Convolutional Neural Networks

Low-light Image Pairs Fusion Method Based on Patch-match

Dual-Focal Camera Hdr Imaging Based On Convolutional Neural Network

Multi-exposure image fusion based on well-exposedness assessment

Ghost Detection And Removal Based On Super-Pixel Grouping In Exposure Fusion

Texture Aware Learning-based Image Fusion Method for Fixed Focal-length Cameras

Multi-Scale Exposure Fusion Via Content Adaptive Edge-Preserving Smoothing Pyramids.

A novel multi-scale fusion framework for detail-preserving low-light image enhancement

Multi-Modal Image Fusion Via Deep Laplacian Pyramid Hybrid Network

Multi-Exposure Fusion of Gray Images Under Low Illumination Based on Low-Rank Decomposition

Enhancing Nighttime Surveillance Video Via Gradient Fusion.

Laplacian and gaussian pyramid based multiscale fusion for nighttime image enhancement

Low-Light Color Imaging via Dual Camera Acquisition.

Infrared and Visible Image Fusion via Sparse Representation and Guided Filtering in Laplacian Pyramid Domain

Detail Enhancement Multi-Exposure Image Fusion Based on Homomorphic Filtering

A Bio-Inspired Multi-Exposure Fusion Framework for Low-light Image Enhancement

Pyramid Fully Convolutional Network for Hyperspectral and Multispectral Image Fusion.

Multi Exposure Fusion for High Dynamic Range Imaging Via Multi-Channel Gradient Tensor

Bilateral fusion low‐light image enhancement with implicit information constraints