Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement

Low-light image enhancement is crucial for accurately interpreting images captured under low-lighting conditions. Existing low-light enhancement methods based on diffusion models have demonstrated effectiveness; however, they suffer from slow training processes and less structured guidance during op...

Full description

Saved in:
Bibliographic Details
Main Authors: Li Li, Jishen Peng, Yingcai Wan
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/3/1604
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850068294029017088
author Li Li
Jishen Peng
Yingcai Wan
author_facet Li Li
Jishen Peng
Yingcai Wan
author_sort Li Li
collection DOAJ
description Low-light image enhancement is crucial for accurately interpreting images captured under low-lighting conditions. Existing low-light enhancement methods based on diffusion models have demonstrated effectiveness; however, they suffer from slow training processes and less structured guidance during optimization. We propose a navel patch-wise-based diffusion model, which introduces the low curvature reverse trajectory to ensure stable parameter optimization and uncertainty guidance in the diffusion training process. Specifically, we randomly select patches of varying sizes from the entire image and apply patch-wise optimization between the generated image and the ground truth to enforce a stable optimization path in the diffusion model. Additionally, within each patch-wise region, an uncertainty network estimates the uncertainty, which is then integrated as a weighting factor in the diffusion process to balance areas of abrupt change in the image. Experimental evaluations on various datasets demonstrate that our method achieves significant improvements, particularly in experiments with real-world images. These results indicate that the proposed patch-wise-based diffusion model enhancements are effective.
format Article
id doaj-art-83f273fdab62405ab0055c5df55d17a3
institution DOAJ
issn 2076-3417
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-83f273fdab62405ab0055c5df55d17a32025-08-20T02:48:06ZengMDPI AGApplied Sciences2076-34172025-02-01153160410.3390/app15031604Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image EnhancementLi Li0Jishen Peng1Yingcai Wan2Faculty of Electrical and Control, Liaoning Technical University, Huludao 125000, ChinaFaculty of Electrical and Control, Liaoning Technical University, Huludao 125000, ChinaSchool of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, ChinaLow-light image enhancement is crucial for accurately interpreting images captured under low-lighting conditions. Existing low-light enhancement methods based on diffusion models have demonstrated effectiveness; however, they suffer from slow training processes and less structured guidance during optimization. We propose a navel patch-wise-based diffusion model, which introduces the low curvature reverse trajectory to ensure stable parameter optimization and uncertainty guidance in the diffusion training process. Specifically, we randomly select patches of varying sizes from the entire image and apply patch-wise optimization between the generated image and the ground truth to enforce a stable optimization path in the diffusion model. Additionally, within each patch-wise region, an uncertainty network estimates the uncertainty, which is then integrated as a weighting factor in the diffusion process to balance areas of abrupt change in the image. Experimental evaluations on various datasets demonstrate that our method achieves significant improvements, particularly in experiments with real-world images. These results indicate that the proposed patch-wise-based diffusion model enhancements are effective.https://www.mdpi.com/2076-3417/15/3/1604diffusion modellow lightuncertainty guidancepatch-wise
spellingShingle Li Li
Jishen Peng
Yingcai Wan
Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement
Applied Sciences
diffusion model
low light
uncertainty guidance
patch-wise
title Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement
title_full Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement
title_fullStr Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement
title_full_unstemmed Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement
title_short Patch-Wise-Based Diffusion Model with Uncertainty Guidance for Low-Light Image Enhancement
title_sort patch wise based diffusion model with uncertainty guidance for low light image enhancement
topic diffusion model
low light
uncertainty guidance
patch-wise
url https://www.mdpi.com/2076-3417/15/3/1604
work_keys_str_mv AT lili patchwisebaseddiffusionmodelwithuncertaintyguidanceforlowlightimageenhancement
AT jishenpeng patchwisebaseddiffusionmodelwithuncertaintyguidanceforlowlightimageenhancement
AT yingcaiwan patchwisebaseddiffusionmodelwithuncertaintyguidanceforlowlightimageenhancement