Fast Enhancement for Non-Uniform Illumination Images using Light-weight CNNs

TL;DR

This paper introduces a lightweight CNN-based method for real-time enhancement of non-uniform illumination images, effectively improving brightness, contrast, and noise reduction with a novel semi-supervised training approach.

Contribution

The paper presents a new ultra-lightweight CNN model and a semi-supervised retouching dataset for efficient non-uniform illumination image enhancement.

Findings

Achieves real-time processing at 50 fps for 0.5 MP images.

Outperforms existing methods in speed and effectiveness.

Handles multiple enhancement aspects simultaneously.

Abstract

This paper proposes a new light-weight convolutional neural network (5k parameters) for non-uniform illumination image enhancement to handle color, exposure, contrast, noise and artifacts, etc., simultaneously and effectively. More concretely, the input image is first enhanced using Retinex model from dual different aspects (enhancing under-exposure and suppressing over-exposure), respectively. Then, these two enhanced results and the original image are fused to obtain an image with satisfactory brightness, contrast and details. Finally, the extra noise and compression artifacts are removed to get the final result. To train this network, we propose a semi-supervised retouching solution and construct a new dataset (82k images) contains various scenes and light conditions. Our model can enhance 0.5 mega-pixel (like 600*800) images in real time (50 fps), which is faster than existing enhancement methods. Extensive experiments show that our solution is fast and effective to deal with non-uniform illumination images.