基于图像自适应增强的低照度目标检测算法

doi:10.19678/j.issn.1000-3428.0068407

计算机工程 ›› 2024, Vol. 50 ›› Issue (10): 352-361. doi: 10.19678/j.issn.1000-3428.0068407

基于图像自适应增强的低照度目标检测算法

王非凡¹^,²^,³, 陈希爱²^,³^,*(), 任卫红⁴, 管宇²^,³^,⁵, 韩志²^,³, 唐延东²^,³

1. 沈阳理工大学自动化与电气工程学院, 辽宁沈阳 110159
2. 中国科学院沈阳自动化研究所机器人学国家重点实验室, 辽宁沈阳 110016
3. 中国科学院机器人与智能制造创新研究院, 辽宁沈阳 110159
4. 哈尔滨工业大学机电工程与自动化学院, 广东深圳 518055
5. 沈阳理工大学信息科学与工程学院, 辽宁沈阳 110159

收稿日期:2023-09-17 出版日期:2024-10-15 发布日期:2024-10-23
通讯作者: 陈希爱
基金资助:
国家自然科学基金(61821005); 中国科学院青年创新促进会项目(2022196); 中国科学院青年创新促进会项目(Y202051); 中国科学院稳定支持基础研究领域青年团队计划(YSBR-041)

Low-illumination Object-Detection Algorithm Based on Image Adaptive Enhancement

WANG Feifan¹^,²^,³, CHEN Xi'ai²^,³^,*(), REN Weihong⁴, GUAN Yu²^,³^,⁵, HAN Zhi²^,³, TANG Yandong²^,³

1. School of Automation and Electrical Engineering, Shenyang Ligong University, Shenyang 110159, Liaoning, China
2. State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, Liaoning, China
3. Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110159, Liaoning, China
4. School of Mechanical Engineering and Automation, Harbin Institute of Technology, Shenzhen 518055, Guangdong, China
5. School of Information Science and Engineering, Shenyang Ligong University, Shenyang 110159, Liaoning, China

Received:2023-09-17 Online:2024-10-15 Published:2024-10-23
Contact: CHEN Xi'ai

摘要/Abstract

摘要：

在低光环境下的检测任务中, 由于低亮度、低对比度以及噪声等不利因素影响, 会存在对目标的漏检、错检等现象。针对此问题, 提出基于图像自适应增强的低照度目标检测算法。将传统图像处理方法与深度学习相结合, 设计图像自适应增强网络, 使用多个可调滤波器通过级联的方式进行结合, 对输入的低光图像进行逐步增强, 各滤波器的调节参数由卷积神经网络根据输入图像的全局信息进行预测。将图像自适应增强网络与YOLOv5目标检测网络相结合进行端到端的联合训练, 使图像增强效果更有利于目标检测。由于在低光目标检测过程中易出现漏检现象, 对通道注意力机制SE-Net进行改进, 设计特征增强网络, 并嵌入到YOLOv5网络中Neck部分的末端, 以减少网络特征融合过程中造成潜在目标特征的信息损失。实验结果表明, 所提算法在真实低光数据集ExDark上的检测精度达到了77.3%, 相较于原始YOLOv5目标检测网络提高了2.1个百分点, 检测速度达到79帧/s, 能够实现实时检测的效果。

关键词: 自适应增强, 低照度, 目标检测, 注意力机制, 联合训练

Abstract:

In the case of detection tasks in low-light environments, owing to the influence of unfavorable factors, such as low brightness, low contrast, and noise, missed and wrong detections can occur. Hence, a low-light object detection algorithm based on image adaptive enhancement is proposed. Combining conventional image processing methods with deep learning, an image adaptive enhancement network is designed, where multiple adjustable filters are combined in cascade to gradually enhance the input low-light image, and the adjustment parameters of each filter are predicted using a convolutional neural network based on the global information of the input image. The adaptive enhancement network is combined with the YOLOv5 object detection network for end-to-end joint training such that the image enhancement effect is more conducive to object detection. As the low-light object detection process is susceptible to missed detection, the channel attention mechanism SE-Net is improved, and a feature enhancement network is designed and embedded into the end of the Neck region of the YOLOv5 network to reduce the loss of information about potential target features caused by the process of fusion of network features. Experimental results show that the proposed algorithm achieves a detection accuracy of 77.3% on the low-light dataset ExDark, which is 2.1 percentage points higher than that afforded by the original YOLOv5 object detection network, and its detection speed reaches 79 frame/s, which affords real-time detection.

Key words: adaptive enhancement, low illumination, object detection, attention mechanisms, joint training

王非凡, 陈希爱, 任卫红, 管宇, 韩志, 唐延东. 基于图像自适应增强的低照度目标检测算法[J]. 计算机工程, 2024, 50(10): 352-361.

WANG Feifan, CHEN Xi'ai, REN Weihong, GUAN Yu, HAN Zhi, TANG Yandong. Low-illumination Object-Detection Algorithm Based on Image Adaptive Enhancement[J]. Computer Engineering, 2024, 50(10): 352-361.

https://www.ecice06.com/CN/Y2024/V50/I10/352

图/表 12

图1 自然光场景和低光场景

Fig.1 Natural light scenes and low light scenes

图2 本文算法框架

Fig.2 The framework of proposed algorithm

图3 自适应增强模块

Fig.3 Adaptive enhancement module

图4 特征增强模块

Fig.4 Feature enhancement module

图5 实验数据集

Fig.5 Experiment datasets

图6 不同模型的检测结果对比

Fig.6 Comparison chart of detection results of different models

图7 图像增强过程

Fig.7 Image enhancement process

图8 YOLOv5与本文模型的P-R曲线

Fig.8 P-R curve of YOLOv5 and proposed model

参考文献 37

1	李雪松, 张锲石, 宋呈群, 等. 自动驾驶场景下的轨迹预测技术综述. 计算机工程, 2023, 49 (5): 1- 11. URL
	LI X S, ZHANG Q S, SONG C Q, et al. Review of trajectory prediction technology in autonomous driving scenes. Computer Engineering, 2023, 49 (5): 1- 11. URL
2	李松江, 耿兰兰, 王鹏. 基于改进Yolov4的车辆目标检测. 计算机工程, 2023, 49 (4): 272- 280. URL
	LI S J, GENG L L, WANG P. Vehicle target detection based on improved Yolov4. Computer Engineering, 2023, 49 (4): 272- 280. URL
3	帅泽群, 李军, 张世义. 适合车载边缘计算的拥挤行人检测算法. 计算机工程与应用, 2023, 59 (4): 156- 164.
	SHUAI Z Q, LI J, ZHANG S Y. Crowded pedestrian detection algorithm suitable for vehicle edge computing. Computer Engineering and Applications, 2023, 59 (4): 156- 164.
4	侯艳丽, 王鑫涛, 魏义仑, 等. 基于IMPSiamCAR孪生网络无人机目标跟踪算法. 计算机应用研究, 2023, 40 (1): 315- 320.
	HOU Y L, WANG X T, WEI Y L, et al. Tracking algorithm of unmanned aerial vehicle targets based on IMPSiamCAR for siamese network. Application Research of Computers, 2023, 40 (1): 315- 320.
5	JIA X Y, ZHU C, LI M Z, et al. LLVIP: a visible-infrared paired dataset for low-light vision[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2021: 3496-3504.
6	ZHAO Z X, BAI H W, ZHANG J S, et al. CDDFuse: correlation-driven dual-branch feature decomposition for multi-modality image fusion[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 5906-5916.
7	LIU J Y, FAN X, HUANG Z B, et al. Target-aware dual adversarial learning and a multi-scenario multi-modality benchmark to fuse infrared and visible for object detection[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans, USA: IEEE Press, 2022: 5802-5811.
8	LIU W Y, REN G F, YU R S, et al. Image-adaptive YOLO for object detection in adverse weather conditions. Artificial Intelligence, 2022, 36 (2): 1792- 1800.
9	CUI Z T, QI G J, GU L, et al. Multitask AET with orthogonal tangent regularity for dark object detection[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2021: 2553-2562.
10	江泽涛, 翟丰硕, 钱艺, 等. 结合特征增强和多尺度感受野的低照度目标检测. 计算机研究与发展, 2023, 60 (4): 903- 915.
	JIANG Z T, ZHAI F S, QIAN Y, et al. Low illumination object detection combined with feature enhancement and multi-scale receptive field. Journal of Computer Research and Development, 2023, 60 (4): 903- 915.
11	HU Y M, HE H, XU C X, et al. Exposure. ACM Transactions on Graphics, 2018, 37 (2): 1- 17.
12	JOCHER G, NISHIMURA K, MINEEVA T. YOLOv5 [EB/OL]. [2023-08-10]. https://github.com/ultralytics/yolov5.
13	VARSHANEYA V, BALASUBRAMANIAN S, GERA D. Res-SE-Net: boosting performance of ResNets by enhancing bridge connections[EB/OL]. [2023-08-10]. http://arxiv.org/pdf/1902.06066.
14	SINGH K, KAPOOR R. Image enhancement using exposure based sub image histogram equalization. Pattern Recognition Letters, 2014, 36, 10- 14.
15	LEE C, LEE C, KIM C S. Contrast enhancement based on layered difference representation of 2D histograms. IEEE Transactions on Image Processing, 2013, 22 (12): 5372- 5384.
16	LAND E H, MCCANN J J. Lightness and retinex theory. Josa, 1971, 61 (1): 1- 11.
17	WEI C, WANG W J, YANG W H, et al. Deep retinex decomposition for low-light enhancement[EB/OL]. [2023-08-10]. http://arxiv.org/abs/1808.04560v1.
18	LI C Y, GUO C L, LOY C C. Learning to enhance low-light image via zero-reference deep curve estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44 (8): 4225- 4238.
19	JIANG Y F, GONG X Y, LIU D, et al. EnlightenGAN: deep light enhancement without paired supervision. IEEE Transactions on Image Processing, 2021, 30, 2340- 2349.
20	LI W, LIANG Z X, MA P, et al. Hausdorff GAN: improving GAN generation quality with Hausdorff metric. IEEE Transactions on Cybernetics, 2022, 52 (10): 10407- 10419.
21	LI W, GU C C, CHEN J L, et al. DW-GAN: toward high-fidelity color-tones of GAN-generated images with dynamic weights. IEEE Transactions on Neural Networks and Learning Systems, 2024, 21, 1- 15.
22	GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: USA: IEEE Press, 2014: 580-587.
23	GIRSHICK R. Fast R-CNN[C]//Proceedings of IEEE International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2015: 1440-1448.
24	REN S, HE K, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39 (6): 1137- 1149.
25	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 21-37.
26	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE Press, 2016: 779-788.
27	REDMON J, FARHADI A. Yolov3: an incremental improvement[EB/OL]. [2023-08-10]. http://arxiv.org/abs/1808.02767v1.
28	BOCHKOVSKIY A, WANG C Y, LIAO H. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. [2023-08-10]. http://arxiv.org/abs/2004.10934v1.
29	王霏, 黄俊, 文洪伟. 基于改进YOLOv3的快速文本检测. 电讯技术, 2022, 62 (1): 130- 137.
	WANG F, HUANG J, WEN H W. Fast text detection based on improved YOLOv3. Telecommunication Engineering, 2022, 62 (1): 130- 137.
30	EVERINGHAM M, ALI ESLAMI S M, VAN GOOL L, et al. The pascal visual object classes challenge: a retrospective. International Journal of Computer Vision, 2015, 111 (1): 98- 136.
31	LOH Y P, CHAN C S. Getting to know low-light images with the exclusively dark dataset. Computer Vision and Image Understanding, 2019, 178, 30- 42.
32	ZHANG Y H, ZHANG J W, GUO X J. Kindling the darkness: a practical low-light image enhancer[C]//Proceedings of the 27th ACM International Conference on Multimedia. New York, USA: ACM Press, 2019: 1632-1640.
33	LIN T Y, MAIRE M, BELONGIE S, et al. Microsoft COCO: common objects in context[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2014: 740-755.
34	CUI Z T, LI K, GU L, et al. You only need 90K parameters to adapt light: a light weight transformer for image enhancement and exposure correction[C]//Proceedings of the 33rd British Machine Vision Conference. London, UK: BMVA Press, 2022: 238.
35	麦锦文, 李浩, 康雁. 基于特征交互结构的弱光目标检测. 计算机工程与应用, 2024, 60 (11): 224- 232.
	MAI J W, LI H, KANG Y. Low-light object detection based on feature interaction structure. Computer Engineering and Applications, 2024, 60 (11): 224- 232.
36	江泽涛, 肖芸, 张少钦, 等. 基于Dark-YOLO的低照度目标检测方法. 计算机辅助设计与图形学学报, 2023, 35 (3): 441- 451.
	JIANG Z T, XIAO Y, ZHANG S Q, et al. Low-illumination object detection method based on dark-YOLO. Journal of Computer-Aided Design & Computer Graphics, 2023, 35 (3): 441- 451.
37	CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[M]. Berlin, Germany: Springer, 2020.

[1]	李俊俊, 董建刚, 李坤. 基于Kubernetes的集群节能策略研究[J]. 计算机工程, 2024, 50(9): 82-91.
[2]	林畅, 郭伟, 任哲聪, 金海波. 基于Transformer的目标跟踪与分割统一算法[J]. 计算机工程, 2024, 50(9): 130-141.
[3]	李泽霖, 吕兆峰, 陈富强, 李克. 基于多跳信息融合的实体对齐模型[J]. 计算机工程, 2024, 50(9): 142-152.
[4]	王汝英, 马嘉骏, 董建强, 刘万龙, 张海涛, 尹凯, 赵博超. 基于MTS-BiGRU-DMHSA的工业负荷预测方法[J]. 计算机工程, 2024, 50(9): 169-178.
[5]	朱凯, 李理, 张彤, 江晟, 别一鸣. 基于Transformer的多阶段运动模糊图像修复网络[J]. 计算机工程, 2024, 50(9): 276-285.
[6]	张天鹏, 韩晶, 吕学强. 基于多任务学习的超分辨率辅助小目标检测[J]. 计算机工程, 2024, 50(9): 304-312.
[7]	郭敏, 张熙涵, 李阳. 融合注意力的教师互一致性半监督医学图像分割[J]. 计算机工程, 2024, 50(9): 313-323.
[8]	曾钰琦, 刘博, 钟柏昌, 钟瑾. 智慧教育下基于改进YOLOv8的学生课堂行为检测算法[J]. 计算机工程, 2024, 50(9): 344-355.
[9]	饶日昕, 王怡文, 曾砺志, 童心恬, 赵海涛. 面向废旧电缆检测的轻量化网络模型[J]. 计算机工程, 2024, 50(8): 22-30.
[10]	李华昱, 张智康, 闫阳, 岳阳. 基于知识图谱增强的领域多模态实体识别[J]. 计算机工程, 2024, 50(8): 31-39.
[11]	王蕾, 党时鹏, 潘丰. 基于卷积神经网络的隐匿性旁路预测模型[J]. 计算机工程, 2024, 50(8): 40-49.
[12]	陈瀚, 赵春蕾, 蒋昊达, 王春东. 基于融合模型与语义网络的App用户意图识别研究[J]. 计算机工程, 2024, 50(8): 50-63.
[13]	王昱婷, 刘志明, 万亚平, 朱涛. 基于可见光与红外图像的弱光条件下目标检测[J]. 计算机工程, 2024, 50(8): 270-281.
[14]	王夙喆, 张雪英, 陈晓玉, 李凤莲, 吴泽林. 基于有效注意力和GAN结合的脑卒中EEG增强算法[J]. 计算机工程, 2024, 50(8): 336-344.
[15]	王宇, 祁琦, 王纯, 许才. 储能变流器信号高精度故障诊断方法[J]. 计算机工程, 2024, 50(8): 389-396.

选择文件类型/文献管理软件名称

选择包含的内容

基于图像自适应增强的低照度目标检测算法

Low-illumination Object-Detection Algorithm Based on Image Adaptive Enhancement

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 37

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于图像自适应增强的低照度目标检测算法

Low-illumination Object-Detection Algorithm Based on Image Adaptive Enhancement

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 37

相关文章 15

编辑推荐

Metrics

本文评价