Recognition-Oriented Low-Light Image Enhancement based on Global and Pixelwise Optimization

TL;DR

This paper introduces a low-light image enhancement technique that improves recognition accuracy by globally adjusting brightness and color, then refining pixel details, without retraining recognition models.

Contribution

The proposed method uniquely combines global and pixelwise optimization modules to enhance recognition performance in low-light images without needing to retrain recognition models.

Findings

Improves recognition accuracy under low-light conditions

Enhances pretrained recognition models without retraining

Effective in real-world low-light scenarios

Abstract

In this paper, we propose a novel low-light image enhancement method aimed at improving the performance of recognition models. Despite recent advances in deep learning, the recognition of images under low-light conditions remains a challenge. Although existing low-light image enhancement methods have been developed to improve image visibility for human vision, they do not specifically focus on enhancing recognition model performance. Our proposed low-light image enhancement method consists of two key modules: the Global Enhance Module, which adjusts the overall brightness and color balance of the input image, and the Pixelwise Adjustment Module, which refines image features at the pixel level. These modules are trained to enhance input images to improve downstream recognition model performance effectively. Notably, the proposed method can be applied as a frontend filter to improve low-light recognition performance without requiring retraining of downstream recognition models. Experimental results demonstrate that our method improves the performance of pretrained recognition models under low-light conditions and its effectiveness.