LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features
Object detection in low-light scenarios has a wide range of applications, but existing algorithms often struggle to preserve the scarce low-level features in dark environments and exhibit limitations in localization accuracy for blurred edges and occluded objects, leading to suboptimal performance....
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10955203/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850177447443562496 |
|---|---|
| author | Wenhao Cai Yajun Chen Xiaoyang Qiu Meiqi Niu Jianying Li |
| author_facet | Wenhao Cai Yajun Chen Xiaoyang Qiu Meiqi Niu Jianying Li |
| author_sort | Wenhao Cai |
| collection | DOAJ |
| description | Object detection in low-light scenarios has a wide range of applications, but existing algorithms often struggle to preserve the scarce low-level features in dark environments and exhibit limitations in localization accuracy for blurred edges and occluded objects, leading to suboptimal performance. To address these challenges, we propose an improved neck structure, SRB-FPN, to achieve fine-grained cross-level semantic alignment and feature fusion, while also optimizing the regression loss function to develop LLD-YOLO, a detector specifically designed for low-light conditions. To enhance the representation of key feature units and dynamically optimize the fusion weights between shallow and deep features, we introduce the SDFBF module. To improve the diversity of receptive fields and strengthen the network’s multi-scale feature capture capability, we incorporate the DBB-C2f module. Furthermore, we integrate the hard-sample focusing property of Focaler IoU with the geometric perception advantages of MPDIoU, proposing Focal MPDIoU Loss to refine the localization of difficult samples and precisely capture bounding box variations. Ultimately, LLD-YOLO achieves an mAP50 of 70.0% on the ExDark dataset, outperforming the baseline by 2.7 percentage points. Extensive experiments on three public datasets, ExDark, NOD, and RTTS, further validate the superior performance of the proposed method in low-light conditions and its strong adaptability to foggy environments. |
| format | Article |
| id | doaj-art-dc2aae3f3ca84375a0c799a8fce99f81 |
| institution | OA Journals |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-dc2aae3f3ca84375a0c799a8fce99f812025-08-20T02:18:58ZengIEEEIEEE Access2169-35362025-01-0113699676997910.1109/ACCESS.2025.355857410955203LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep FeaturesWenhao Cai0https://orcid.org/0009-0009-5792-1373Yajun Chen1https://orcid.org/0009-0006-2494-1878Xiaoyang Qiu2https://orcid.org/0009-0002-5446-2887Meiqi Niu3Jianying Li4https://orcid.org/0009-0000-7690-7563School of Electronic Information Engineering, China West Normal University, Nanchong, ChinaSchool of Electronic Information Engineering, China West Normal University, Nanchong, ChinaSchool of Electronic Information Engineering, China West Normal University, Nanchong, ChinaSchool of Electronic Information Engineering, China West Normal University, Nanchong, ChinaSchool of Electronic Information Engineering, China West Normal University, Nanchong, ChinaObject detection in low-light scenarios has a wide range of applications, but existing algorithms often struggle to preserve the scarce low-level features in dark environments and exhibit limitations in localization accuracy for blurred edges and occluded objects, leading to suboptimal performance. To address these challenges, we propose an improved neck structure, SRB-FPN, to achieve fine-grained cross-level semantic alignment and feature fusion, while also optimizing the regression loss function to develop LLD-YOLO, a detector specifically designed for low-light conditions. To enhance the representation of key feature units and dynamically optimize the fusion weights between shallow and deep features, we introduce the SDFBF module. To improve the diversity of receptive fields and strengthen the network’s multi-scale feature capture capability, we incorporate the DBB-C2f module. Furthermore, we integrate the hard-sample focusing property of Focaler IoU with the geometric perception advantages of MPDIoU, proposing Focal MPDIoU Loss to refine the localization of difficult samples and precisely capture bounding box variations. Ultimately, LLD-YOLO achieves an mAP50 of 70.0% on the ExDark dataset, outperforming the baseline by 2.7 percentage points. Extensive experiments on three public datasets, ExDark, NOD, and RTTS, further validate the superior performance of the proposed method in low-light conditions and its strong adaptability to foggy environments.https://ieeexplore.ieee.org/document/10955203/Keywords object detectionlow-light scenariosYOLOdynamic feature fusion |
| spellingShingle | Wenhao Cai Yajun Chen Xiaoyang Qiu Meiqi Niu Jianying Li LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features IEEE Access Keywords object detection low-light scenarios YOLO dynamic feature fusion |
| title | LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features |
| title_full | LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features |
| title_fullStr | LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features |
| title_full_unstemmed | LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features |
| title_short | LLD-YOLO: A Low-Light Object Detection Algorithm Based on Dynamic Weighted Fusion of Shallow and Deep Features |
| title_sort | lld yolo a low light object detection algorithm based on dynamic weighted fusion of shallow and deep features |
| topic | Keywords object detection low-light scenarios YOLO dynamic feature fusion |
| url | https://ieeexplore.ieee.org/document/10955203/ |
| work_keys_str_mv | AT wenhaocai lldyoloalowlightobjectdetectionalgorithmbasedondynamicweightedfusionofshallowanddeepfeatures AT yajunchen lldyoloalowlightobjectdetectionalgorithmbasedondynamicweightedfusionofshallowanddeepfeatures AT xiaoyangqiu lldyoloalowlightobjectdetectionalgorithmbasedondynamicweightedfusionofshallowanddeepfeatures AT meiqiniu lldyoloalowlightobjectdetectionalgorithmbasedondynamicweightedfusionofshallowanddeepfeatures AT jianyingli lldyoloalowlightobjectdetectionalgorithmbasedondynamicweightedfusionofshallowanddeepfeatures |