Low-light Image Enhancement Using Cone-Rod Coupling.

Area of Science:

• Computer Vision
• Biomedical Engineering
• Image Processing

Background:

• Low-light conditions degrade image quality, causing insufficient brightness, noise, and poor color contrast.
• This degradation significantly impacts the performance of computer vision systems.
• Existing enhancement methods often struggle to balance brightness, color, and detail preservation.

Purpose of the Study:

• To develop a novel low-light image enhancement method inspired by the human visual system's cone-rod coupling mechanism.
• To address the challenges of insufficient brightness, noise, and low color contrast in low-light images.
• To improve the performance of computer vision applications operating in low-light environments.

Main Methods:

• Designed two lightweight sub-networks based on cone and rod cell properties for brightness adjustment and color correction using deep gamma correction.
• Incorporated an efficient Transformer network for post-processing to enhance details and reduce noise.
• Employed quantitative and qualitative evaluations to compare the proposed method against state-of-the-art techniques.

Main Results:

• The proposed bio-inspired method demonstrated superior performance compared to existing state-of-the-art low-light image enhancement techniques.
• Achieved more natural color fidelity and a balanced brightness distribution in enhanced images.
• Effectively enhanced image details and filtered noise, leading to improved visual quality.

Conclusions:

• The novel low-light image enhancement method, inspired by the cone-rod coupling mechanism, offers significant improvements in image quality.
• The approach provides a promising solution for computer vision tasks requiring high-quality images in low-light conditions.
• This bio-inspired technique offers a new direction for developing advanced image processing algorithms.