改进YOLOv11的低光照目标检测方法研究

doi:10.3778/j.issn.1002-8331.2506-0263

摘要/Abstract

摘要： 针对低光照复杂场景下检测的性能瓶颈，提出一种改进YOLOv11的低光照目标检测算法ELS-YOLO（enhanced low-light scene-YOLO）。引入EfficientNetV2替换主干网络，简化模型结构的同时，提升局部细节和全局背景的复杂交互。提出了一种频域感知模块，进一步提升网络对图像细节的感知效果，帮助模型更好地识别低光照条件下的目标轮廓和暗部细节。设计了一种基于灵慧阶梯增益分配策略的损失函数，在提升光照和噪声变化的适应能力基础上持续提升检测精度。算法的消融实验和对比实验在ExDark、NOD（night object detection）和VOC2012数据集上进行，结果表明，在三个数据集上，提出的算法相较于基线模型在mAP@0.5和mAP@0.5：0.95指标上分别提升了4.2和2.6个百分点、2.4和0.7个百分点、0.7和0.3个百分点。实验结果验证了该算法在低光照目标检测场景中的有效性。

关键词: 低光照, 目标检测, YOLOv11, 频域感知

Abstract: For the performance bottleneck in detecting low-light complex scenarios, this paper proposes a low light target detection algorithm ELS-YOLO (enhanced low-light scene-YOLO) based on improved YOLOv11. Firstly, EfficientNetV2 is introduced to replace the backbone network, simplifying the model structure while enhancing the complex interaction between local details and global background. Secondly, a frequency domain perception module is proposed to further improve the network’s perception of image details, helping the model better identify the contours and dark details of objects under low-light conditions. Finally, a loss function based on the wise stepwise gain allocation strategy is designed, which continuously improves detection accuracy while enhancing the adaptability to light and noise changes. The ablation and comparative experiments are carried out on the ExDark, NOD (night object detection) and the VOC2012 datasets. The results demonstrate that on the three datasets, the proposed algorithm outperforms the baseline model by 4.2 and 2.6 percentage points, 2.4 and 0.7 percentage points, and 0.7 and 0.3 percentage points in terms of mAP@0.5 and mAP@0.5：0.95, respectively. These findings validate the effectiveness of the algorithm in low-light object detection scenarios.

Key words: low light, object detection, YOLOv11, frequency domain sensing

李俊林, 张雪松, 宋存利, 李光宇. 改进YOLOv11的低光照目标检测方法研究[J]. 计算机工程与应用, 2026, 62(5): 106-119.

LI Junlin, ZHANG Xuesong, SONG Cunli, LI Guangyu. Research on Low-light Object Detection Method of Improved YOLOv11[J]. Computer Engineering and Applications, 2026, 62(5): 106-119.

参考文献

[1] 李艳，张科，李明辉.微光场景下基于改进自适应Retinex的视觉SLAM算法[J/OL].计算机工程与应用（2025-05-21）[2025-05-25].https：//link.cnki.net/urlid/11.2127.TP.20250520. 1646.021.
LI Y，ZHANG K，LI M H.Visual SLAM algorithm based on improved adaptive Retinex in low light scene[J/OL].Computer Engineering and Applications（2025-05-21）[2025-05-25].https：//link.cnki.net/urlid/11.2127.TP.20250520.1646.021.
[2] 陈谢发，李敏，赵晶钰，等.基于特征融合增强的低光照目标检测算法[J].计算机工程与应用，2025，61（22）：235-244.
CHEN X F，LI M，ZHAO J Y，et al.Low light target detection algorithm based on feature fusion enhancement[J].Computer Engineering and Applications，2025，61（22）：235-244.
[3] QIU Y S，LU Y Y，WANG Y T，et al.IDOD-YOLOV7：image-dehazing YOLOV7 for object detection in low-light foggy traffic environments[J].Sensors，2023，23（3）：1347.
[4] WEI C，WANG W J，YANG W H，et al.Deep retinex decomposition for low-light enhancement[EB/OL].[2025-05-25].https：//arxiv.org/abs/1808.04560.
[5] GUO C L，LI C Y，GUO J C，et al.Zero-reference deep curve estimation for low-light image enhancement[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2020：1777-1786.
[6] JIANG Y F，GONG X Y，LIU D，et al.EnlightenGAN：deep light enhancement without paired supervision[J].IEEE Transactions on Image Processing，2021，30：2340-2349.
[7] MA L，MA T Y，LIU R S，et al.Toward fast，flexible，and robust low-light image enhancement[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2022：5627-5636.
[8] CAI Y H，BIAN H，LIN J，et al.RetinexFormer：one-stage retinex-based transformer for low-light image enhancement[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2024：12470-12479.
[9] REN S Q，HE K M，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（6）：1137-1149.
[10] FENG X，ZENG J X，WANG S P，et al.Toward highly efficient semantic-guided machine vision for low-light object detection[C]//Proceedings of the 35th British Machine Vision Conference，2024：25-88.
[11] WANG K，LIU M Z.Object recognition at night scene based on DCGAN and faster R-CNN[J].IEEE Access，2020，8：193168-193182.
[12] RADFORD A，METZ L，CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks[EB/OL].[2025-05-25].https：//arxiv.org/abs/1511.06434.
[13] XU X，WANG S，WANG Z，et al.Exploring image enhancement for salient object detection in low light images[J].ACM Transactions on Multimedia Computing，Communications，and Applications，2021，17（1S）：1-19.
[14] ZHAO D W，SHAO F M，ZHANG S，et al.Advanced object detection in low-light conditions：enhancements to YOLOv7 framework[J].Remote Sensing，2024，16（23）：4493.
[15] WANG C Y，BOCHKOVSKIY A，LIAO H M.YOLOv7：trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2023：7464-7475.
[16] HUI Y M，WANG J，LI B.WSA-YOLO：weak-supervised and adaptive object detection in the low-light environment for YOLOV7[J].IEEE Transactions on Instrumentation and Measurement，2024，73：2507012.
[17] KOOHESTANI F，KARIMI N，SAMAVI S.Revealing shadows：low-light image enhancement using self-calibrated illumination[C]//Proceedings of the 2024 32nd International Conference on Electrical Engineering.Piscataway：IEEE，2024：1-7.
[18] KIM S J，KIM M J.Human action recognition using YOLOv11 ultralytics：a comprehensive study for real-time applications[J].Journal of Advanced Artificial Intelligence，Engineering and Technology，2025，1（2）：1-14.
[19] TAN M X，LE Q V.EfficientNetV2：smaller models and faster training[C]//Proceedings of the International Conference on Machine Learning，2021：10096-10106.
[20] AL SOBBAHI R，TEKLI J.Comparing deep learning models for low-light natural scene image enhancement and their impact on object detection and classification：overview，empirical evaluation，and challenges[J].Signal Processing：Image Communication，2022，109：116848.
[21] KHANAM R，HUSSAIN M.YOLOv11：an overview of the key architectural enhancements[EB/OL].[2025-05-26].https：//arxiv.org/abs/2410.17725.
[22] DU S J，ZHANG B F，ZHANG P，et al.An improved bounding box regression loss function based on CIOU loss for multi-scale object detection[C]//Proceedings of the 2021 IEEE 2nd International Conference on Pattern Recognition and Machine Learning.Piscataway：IEEE，2021：92-98.
[23] LIU C，WANG K G，LI Q，et al.Powerful-IoU：more straightforward and faster bounding box regression loss with a nonmonotonic focusing mechanism[J].Neural Networks，2024，170：276-284.
[24] TONG Z J，CHEN Y H，XU Z W，et al.Wise-IoU：bounding box regression loss with dynamic focusing mechanism[EB/OL].[2025-05-26].https：//arxiv.org/abs/2301.10051.
[25] FU Y，HONG Y，CHEN L W，et al.LE-GAN：unsupervised low-light image enhancement network using attention module and identity invariant loss[J].Knowledge-Based Systems，2022，240：108010.
[26] STEVENS E，ANTIGA L，VIEHMANN T.Deep learning with PyTorch：build，train，and tune neural networks using PyThon tools[M].Greenwich：Manning Publication Co，2020.
[27] LOH Y P，CHAN C S.Getting to know low-light images with the exclusively dark dataset[J].Computer Vision and Image Understanding，2019，178：30-42.
[28] MORAWSKI I，CHEN Y，LIN Y S，et al.NOD：taking a closer look at detection under extreme low-light conditions with night object detection dataset[EB/OL].[2025-05-26].https：//arxiv.org/abs/2110.10364.
[29] Everingham M，Winn J.The PASCAL visual object classes challenge 2012 （VOC2012） development kit[EB/OL].[2025-05-27].https：//storage.googleapis.com/kaggle-media/compe-titions/SIIM2021/VOC2012_doc.pdf.
[30] JUBA B，LE H S.Precision-recall versus accuracy and the role of large data sets[C]//Proceedings of the 33rd AAAI Conference on Artificial Intelligence and 31st Innovative Applications of Artificial Intelligence Conference and 9th AAAI Symposium on Educational Advances in Artificial Intelligence.New York：ACM，2019：4039-4048.
[31] LI J W，YU H W，CHEN J S，et al.A2RNet：adversarial attack resilient network for robust infrared and visible image fusion[C]//Proceedings of the 39th AAAI Conference on Artificial Intelligence，2025：4770-4778.
[32] SU J N，GAN M，CHEN G Y，et al.High-similarity-pass att-ention for single image super-resolution[J].IEEE Transactions on Image Processing，2024，33：610-624.
[33] LU L P，XIONG Q，XU B R，et al.MixDehazeNet：mix structure block for image dehazing network[C]//Proceedings of the 2024 International Joint Conference on Neural Networks.Piscataway：IEEE，2024：1-10.
[34] JIANG X Y，ZHANG X H，GAO N，et al.When fast Fourier transform meets transformer forImage restoration[C]//Proceedings of the 18th European Conference on Computer Vision.Cham：Springer，2025：381-402.
[35] ZHENG Z H，WANG P，REN D W，et al.Enhancing geometric factors in model learning and inference for object detection and instance segmentation[J].IEEE Transactions on Cybernetics，2022，52（8）：8574-8586.
[36] REZATOFIGHI H，TSOI N，GWAK J，et al.Generalized intersection over union：a metric and a loss for bounding box regression[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2020：658-666.
[37] ZHANG H，ZHANG S J.Shape-IoU：more accurate metric considering bounding box shape and scale[EB/OL].[2025-05-28].https：//arxiv.org/abs/2312.17663.
[38] 蒋畅江，何旭颖，向杰.LOL-YOLO：融合多注意力机制的低照度目标检测[J].计算机工程与应用，2024，60（24）：177-187.
JIANG C J，HE X Y，XIANG J.LOL-YOLO：low-light object detection incorporating multiple attention mechanisms[J].Computer Engineering and Applications，2024，60（24）：177-187.
[39] GEVORGYAN Z.SIoU loss：more powerful learning for bounding box regression[EB/OL].[2025-05-28].https：//arxiv.org/abs/2205.12740.
[40] ZHENG Z H，WANG P，LIU W，et al.Distance-IoU loss：faster and better learning for bounding box regression[C]//Proceedings of the 34th AAAI Conference on Artificial Intelligence，2020：12993-13000.
[41] LI C Y，LI L L，JIANG H L，et al.YOLOv6：a single-stage object detection framework for industrial applications[EB/OL].[2025-05-28].https：//arxiv.org/abs/2209.02976.
[42] WANG C Y，YEH I H，MARK LIAO H Y.YOLOv9：lear-ning what you want to learn using programmable gradient information[C]//Proceedings of the 18th European Conference on Computer Vision.Cham：Springer，2025：1-21.
[43] WANG A，CHEN H，LIU L H，et al.YOLOv10：real-time end-to-end object detection[C]//Advances in Neural Information Processing Systems 38，2024：107984-108011.
[44] TIAN Y J，YE Q X，DOERMANN D.YOLOv12：attention-centric real-time object detectors[EB/OL].[2025-05-29].https：//arxiv.org/abs/2502.12524.
[45] CARION N，MASSA F，SYNNAEVE G，et al.End-to-end object detection with transformers[C]//Proceedings of the 16th European Conference on Computer Vision.Cham：Springer，2020：213-229.
[46] CUI Z T，QI G J，GU L，et al.Multitask AET with orthogonal tangent regularity for dark object detection[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2022：2533-2542.
[47] CHATTOPADHAY A，SARKAR A，HOWLADER P，et al.Grad-CAM++：generalized gradient-based visual explanations for deep convolutional networks[C]//Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision.Piscataway：IEEE，2018：839-847.