留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

夜间动物图像自监督学习增强与检测方法

王驰 沈晨 黄庆 张国峰 卢汉 陈金波

王驰, 沈晨, 黄庆, 张国峰, 卢汉, 陈金波. 夜间动物图像自监督学习增强与检测方法[J]. 中国光学(中英文). doi: 10.37188/CO.2024-0011
引用本文: 王驰, 沈晨, 黄庆, 张国峰, 卢汉, 陈金波. 夜间动物图像自监督学习增强与检测方法[J]. 中国光学(中英文). doi: 10.37188/CO.2024-0011
WANG Chi, SHEN Chen, HUANG Qing, ZHANG Guo-feng, LU Han, CHEN Jin-bo. Self-supervised learning enhancement and detection methods for nocturnal animal images[J]. Chinese Optics. doi: 10.37188/CO.2024-0011
Citation: WANG Chi, SHEN Chen, HUANG Qing, ZHANG Guo-feng, LU Han, CHEN Jin-bo. Self-supervised learning enhancement and detection methods for nocturnal animal images[J]. Chinese Optics. doi: 10.37188/CO.2024-0011

夜间动物图像自监督学习增强与检测方法

doi: 10.37188/CO.2024-0011
基金项目: 国家自然科学基金项目(No. 62175144),北京市航空智能遥感装备工程技术研究中心开放基金课题(No. AIRSE20233)
详细信息
    作者简介:

    王 驰(1982—),男,河南周口人,博士(后),教授,2009 年于天津大学获得博士学位,现为上海大学机电工程与自动化学院教师,主要从事精密测试技术及仪器等方面的研究。 E-mail:wangchi@shu.edu.cn

    陈金波(1980—),男,内蒙古呼和浩特人,博士,工程师,2002年于上海大学获得硕士学位,2014年于上海大学获得博士学位,现为上海大学机电工程与自动化学院教师,主要从事机器视觉与检测方面的研究。E-mail:jbchen@shu.edu.cn

  • 中图分类号: TP394.1;TH691.9

Self-supervised learning enhancement and detection methods for nocturnal animal images

Funds: Supported by National Natural Science Foundation of China (No.62175144),Beijing Engineering Research Center of Aerial Intelligent Remote Sensing Equipments Fund (No. AIRSE20233)
More Information
  • 摘要:
    目的 

    为了解决动物夜间实时监测所面临的图像曝光度低、对比度低、特征提取困难等问题,通过研究轻量化自监督深度神经网络Zero-Denoise和改进型YOLOv8模型,来进行夜间动物目标的图像增强与精准识别。

    方法 

    首先,通过轻量化的PDCE-Net进行第一阶段快速增强。提出了一个新的光照损失函数,并利用参数可调的Gamma校正的原图与快速增强图,在基于Retinex原理和最大熵理论的PRED-Net中进行第二阶段的重增强。然后,改进YOLOv8模型,并对重增强后的图像进行目标识别。最后,在LOL数据集与自建动物数据集进行实验分析,验证Zero-Denoise网络和改进型YOLOv8模型用于夜间动物目标监测的改善。

    结果 

    试验结果显示,Zero-Denoise 网络在 LOL 数据集上的PSNR、SSIM与MAE指标达到28.53、0.76、26.15,结合改进型 YOLOv8 在自建动物数据集上比 YOLOv8 基线模型的mAP值提升7.1%。

    结论 

    使用 Zero-Denoise和改进型YOLOv8能获得夜间动物目标的良好的图像质量,可用于夜间动物目标精确监测方法的进一步研究。

     

  • 图 1  部分卷积PConv原理图

    Figure 1.  schematic diagram of PConv

    图 2  Zero-Denoise网络整体结构

    Figure 2.  Overall Structure of Zero-Denoise Network

    图 3  SA注意力机制原理图

    Figure 3.  Schematic diagram of Shuffle Attention mechanism

    图 4  RepGFPN整体结构

    Figure 4.  Overall structure of RepGPFN

    图 5  改进YOLOv8网络整体结构

    Figure 5.  Overall structure of improved YOLOv8

    图 6  数据集各类动物种类

    Figure 6.  Various animal species in the dataset

    图 7  各类先进图像增强算法对比

    Figure 7.  Comparison of various advanced image enhancement algorithms

    图 8  增强与未增强的目标检测可视化结果

    Figure 8.  Enhanced and unenhanced target detection visualization results

    表  1  不同算法评价指标结果

    Table  1.   Results of different algorithms for evaluating indicators

    算法PSNRSSIMMAERuntime(s)
    无监督
    自监督
    Zero-DCE ++27.930.572544.93930.0008
    SCI27.900.525448.75500.0006
    Zero-Denoise(ours)28.530.766526.15160.0181
    有监督StableLLVE27.920.737332.47890.5900
    URetinex-Net28.450.833221.15620.0367
    Retinexnet28.060.425032.01740.1200
    MBLLEN28.040.724731.24988.5633
    Zero-Denoise(ours)28.530.766526.15160.0181
    下载: 导出CSV

    表  2  改进YOLOv8消融实验结果

    Table  2.   Experimental results of improved YOLOv8 ablation

    算法改进主干改进颈部改进损失PRmAP@0.5mAP@0.5:0.95
    YOLOv872.370.880.749.7
    A73.3(+1.0%)72.3(+1.5%)81.2(+0.5%)50.3(+0.6%)
    B75.0(+2.7%)74.6(+3.8%)81.4(+0.7%)50.4(+0.7%)
    C73.3(+0.0%)72.3(+1.5%)80.9(+0.2%)50.1(+0.4%)
    D76.0(+3.7%)74.9(+4.1%)81.9(+1.2%)51.3(+1.4%)
    Ours77.9(5.6%)75.2(+5.6%)82.2(+1.5%)51.7+(2.0%)
    下载: 导出CSV

    表  3  不同算法对改进YOLOv8的影响

    Table  3.   The impact of different algorithms on improved YOLOv8

    算法RabbitBirdChickenMouseDuckmAP@0.5
    Im-YOLOv894.086.562.280.363.582.2
    URetinex-Net+Im-YOLOv894.588.073.591.966.283.5(+1.3%)
    StableLLVE+Im-YOLOv893.590.671.682.264.582.5(+0.3%)
    SCI+Im-YOLOv894.791.172.093.269.584.7(+2.5%)
    Retinexnet+Im-YOLOv889.276.660.968.558.778.5(-3.7%)
    MBLLEN+Im-YOLOv894.290.872.589.065.583.1(+1.1%)
    Zero-Denoise+Im-YOLOv895.091.774.697.476.087.8(+5.6%)
    下载: 导出CSV

    表  4  增强与未增强的目标检测可视化结果对比

    Table  4.   Comparison of enhanced and unenhanced target detection visualization results

    组别 应检个数/实检个数 检测精度 结论
    a1鸽类增强后检测结果 3/3 0.49/0.79/0.89 未增强图像出现漏检数量1,阴影处鸽类未检出
    a2鸽类未增强检测结果 3/2 0.00/0.81/0.88
    b1鸡类增强后检测结果 6/6 0.55/0.80/0.81/0.85/0.57/0.46 未增强图像出现漏检数量2,角落处和被遮挡的鸡类未检出
    b1鸡类未增强检测结果 6/4 0.00/0.84/0.89/0.83/0.85/0.00
    c1鸭类增强后检测结果 3/3 0.50/0.86/0.81 未增强图像出现漏检数量1,深处的鸭类未检出
    c2鸭类未增强检测结果 3/2 0.00/0.71/0.68
    d1鸭类增强后检测结果 4/4 0.27/0.66/0.82/0.36 未增强图像出现漏检数量2,左侧和右侧阴影内鸭类未检出
    d2鸭类未增强检测结果 4/2 0.00/0.31/0.88/0.00
    e1鼠类增强后检测结果 4/4 0.87/0.82/0.88/0.80 增强图像中鼠类检测精度提升明显从0.27、0.26提升至0.82、0.80
    e2鼠类未增强检测结果 4/4 0.83/0.27/0.87/0.26
    f1兔类增强后检测结果 1/1 0.90 增强图像中兔类检测精度提升明显从0.77提升至0.90
    f2兔类未增强检测结果 1/1 v 0.77
    下载: 导出CSV
  • [1] QI Y L, YANG ZH, SUN W H, et al. A comprehensive overview of image enhancement techniques[J]. Archives of Computational Methods in Engineering, 2022, 29(1): 583-607. doi: 10.1007/s11831-021-09587-6
    [2] 刘彦磊, 李孟喆, 王宣宣. 轻量型YOLOv5s车载红外图像目标检测[J]. 中国光学(中英文),2023,16(5):1045-1055. doi: 10.37188/CO.2022-0254

    LIU Y L, LI M ZH, WANG X X. Lightweight YOLOv5s vehicle infrared image target detection[J]. Chinese Optics, 2023, 16(5): 1045-1055. (in Chinese). doi: 10.37188/CO.2022-0254
    [3] HU H F. Illumination invariant face recognition based on dual-tree complex wavelet transform[J]. IET Computer Vision, 2015, 9(2): 163-173. doi: 10.1049/iet-cvi.2013.0342
    [4] MUNIAN Y, MARTINEZ-MOLINA A, MISERLIS D, et al. Intelligent system utilizing HOG and CNN for thermal image-based detection of wild animals in nocturnal periods for vehicle safety[J]. Applied Artificial Intelligence, 2022, 36(1): 2031825. doi: 10.1080/08839514.2022.2031825
    [5] MURUGAN R A, SATHYABAMA B. Object detection for night surveillance using Ssan dataset based modified Yolo algorithm in wireless communication[J]. Wireless Personal Communications, 2023, 128(3): 1813-1826. doi: 10.1007/s11277-022-10020-9
    [6] BHATT D, PATEL C, TALSANIA H, et al. CNN variants for computer vision: history, architecture, application, challenges and future scope[J]. Electronics, 2021, 10(20): 2470. doi: 10.3390/electronics10202470
    [7] 任凤雷, 周海波, 杨璐, 等. 基于双注意力机制的车道线检测[J]. 中国光学(中英文),2023,16(3):645-653. doi: 10.37188/CO.2022-0033

    REN F L, ZHOU H B, YANG L, et al. Lane detection based on dual attention mechanism[J]. Chinese Optics, 2023, 16(3): 645-653. (in Chinese). doi: 10.37188/CO.2022-0033
    [8] LI CH Y, GUO J CH, PORIKLI F, et al. LightenNet: a Convolutional Neural Network for weakly illuminated image enhancement[J]. Pattern Recognition Letters, 2018, 104: 15-22. doi: 10.1016/j.patrec.2018.01.010
    [9] DING X, HU R M. Learning to see faces in the dark[C]. 2020 IEEE International Conference on Multimedia and Expo (ICME), IEEE, 2020: 1-6.
    [10] LI CH Y, GUO CH L, CHEN CH L. Learning to enhance low-light image via zero-reference deep curve estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 40(8): 4225-4238.
    [11] 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. doi: 10.1109/TIP.2021.3051462
    [12] 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. doi: 10.1016/j.knosys.2021.108010
    [13] WANG R J, JIANG B, YANG CH, et al. MAGAN: Unsupervised low-light image enhancement guided by mixed-attention[J]. Big Data Mining and Analytics, 2022, 5(2): 110-119. doi: 10.26599/BDMA.2021.9020020
    [14] MONAKHOVA K, RICHTER S R, WALLER L, et al. Dancing under the stars: video denoising in starlight[C]. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2022: 16220-16230.
    [15] CHEN J R, KAO S H, HE H, et al. Run, Don't walk: chasing higher FLOPS for faster neural networks[C]. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2023: 12021-12031.
    [16] WEI CH, WANG W J, YANG W H, et al. Deep retinex decomposition for low-light enhancement[C]. British Machine Vision Conference 2018, BMVA Press, 2018: 155.
    [17] ZHANG Y, DI X G, ZHANG B, et al. Self-supervised low light image enhancement and denoising[J]. arXiv preprint arXiv: 2103.00832, 2021. (查阅网上资料, 不确定类型及格式是否正确, 请确认) .
    [18] ZHANG Q L, YANG Y B. SA-Net: Shuffle attention for deep convolutional neural networks[C]. 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, 2021: 2235-2239.
    [19] XU X ZH, JIANG Y Q, CHEN W H, et al. DAMO-YOLO: a report on real-time object detection design[J]. arXiv preprint arXiv: 2211.15444, 2022. (查阅网上资料, 不确定类型及格式是否正确, 请确认) .
    [20] TONG Z J, CHEN Y H, XU Z W, et al. Wise-IoU: Bounding box regression loss with dynamic focusing mechanism[J]. arXiv preprint arXiv: 2301.10051, 2023. (查阅网上资料, 不确定类型及格式是否正确, 请确认) .
    [21] WU W H, WENG J, ZHANG P P, et al. URetinex-Net: Retinex-based deep unfolding network for low-light image enhancement[C]. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2022: 5891-5900.
    [22] ZHANG F, LI Y, YOU SH D, et al. Learning temporal consistency for low light video enhancement from single images[C]. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2021: 4965-4974.
    [23] MA L, MA T Y, LIU R SH, et al. Toward fast, flexible, and robust low-light image enhancement[C]. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2022: 5627-5636.
    [24] LV F F, LU F, WU J H, et al. MBLLEN: Low-light image/video enhancement using CNNs[C]. British Machine Vision Conference 2018, BMVA Press, 2018: 220.
  • 加载中
图(8) / 表(4)
计量
  • 文章访问数:  46
  • HTML全文浏览量:  38
  • PDF下载量:  5
  • 被引次数: 0
出版历程
  • 网络出版日期:  2024-06-17

目录

    /

    返回文章
    返回