面向废旧电缆检测的轻量化网络模型

doi:10.19678/j.issn.1000-3428.0068537

摘要/Abstract

摘要：

目前我国废旧电缆回收主要依靠人工分拣, 存在耗时耗力、准确性低的问题。为了更好地将模型部署至小型化硬件设备并提高检测实时性, 提出基于改进YOLOv5s的废旧电缆检测网络模型。首先, 将主干网络中的标准卷积模块替换为轻量化的幻象(Ghost)模块以减小网络的复杂度, 并且在快速空间金字塔池化(SPPF)模块前引入卷积块注意力模块(CBAM), 提高了特征提取和融合的效率。其次, 将网络中Neck部分的C3模块结合有效通道注意力模块(ECA), 实现了跨通道的信息交互, 提高了网络特征融合能力。最后, 在损失函数的计算部分使用Wise-交并比(WIoU)作为新的边界框损失函数以提升回归效果, 加快模型的收敛速度。实验结果表明: 改进模型的平均检测精度为96.3%, 相比单点多框检测器(SSD)提高了1.2个百分点; 参数量为5.15×10⁶, 相比YOLOv5s模型减少了27.0%;在小型嵌入式设备LubanCat-1上的推理速度达到8.49帧/s, 具有良好的实时性, 适用于废旧电缆的实时检测与分类。

关键词: 废旧电缆检测, YOLOv5s模型, 轻量化, 注意力机制, 嵌入式设备

Abstract:

Currently, waste cable recycling in China relies primarily on manual sorting, which is time-consuming, labor-intensive, and inaccurate. To deploy the model on small embedded devices more effectively and improve real-time detection, an improved lightweight network model based on YOLOv5s for waste cable detection is proposed herein. First, standard convolution modules in the Backbone of the network are replaced with lightweight Ghost modules to reduce network complexity, and a Convolutional Block Attention Module (CBAM) is introduced before a Fast Spatial Pyramid Pooling (SPPF) module to enhance feature extraction and fusion efficiency. Second, the C3 module in the neck of the network is combined with an Effective Channel Attention (ECA) module to facilitate inter-channel information interaction and enhance the network's feature fusion capability. Finally, Wise Intersection over Union (WIoU) is utilized as a new bounding box loss function to improve the regression effect and accelerate the model convergence speed. The experimental results demonstrate that the improved model achieves an average detection accuracy of 96.3%, which is 1.2 percentage points higher than that of the Single Shot multibox Detector (SSD). There are 5.15×10⁶ parameters in the proposed model, which is a 27.0% reduction compared with the YOLOv5s model. The inference speed on the small embedded device, LubanCat-1, reaches 8.49 frame/s, indicating excellent real-time performance and suitability for real-time detection and classification of waste cables.

Key words: waste cable detection, YOLOv5s model, lightweight, attention mechanism, embedded device

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.

Rixin RAO, Yiwen WANG, Lizhi ZENG, Xintian TONG, Haitao ZHAO. Lightweight Network Model for Waste Cable Detection[J]. Computer Engineering, 2024, 50(8): 22-30.

https://www.ecice06.com/CN/Y2024/V50/I8/22

图/表 15

图1 改进的YOLOv5s网络结构

Fig.1 Structure of improved YOLOv5s network

图2 Ghost模块结构

Fig.2 Structure of Ghost module

图3 GhostBottleneck与C3Ghost结构

Fig.3 Structure of GhostBottleneck and C3Ghost

图4 CBAM模块结构

Fig.4 Structure of CBAM module

图5 ECA模块结构

Fig.5 Structure of ECA module

图6 C3ECA模块结构

Fig.6 Structure of C3ECA module

图7 废旧电缆数据集分类

Fig.7 Classification of waste cable dataset

图8 数据加强处理

Fig.8 Data augmentation processing

图9 检测结果可视化对比

Fig.9 Visual comparison of detection results

图10 不同运行环境对比实验结果

Fig.10 Comparative experimental results of different runtime environments

图11 损失曲线对比

Fig.11 Comparison of loss curves

参考文献 26

1	张楠. 新能源产业发展背景下我国铜资源供需现状与趋势. 中国矿业, 2023, 32(6): 2- 9. URL
	ZHANG N. Analysis of supply and demand status and trend of copper resources in China under development background of new energy industry. China Mining Magazine, 2023, 32(6): 2- 9. URL
2	文博杰, 代涛, 韩中奎, 等. 中国铜资源在用存量与二次供应潜力. 地球学报, 2023, 44(2): 325- 332. URL
	WEN B J, DAI T, HAN Z K, et al. Copper in-use stock and recycling potential in China. Acta Geoscientica Sinica, 2023, 44(2): 325- 332. URL
3	REN S Q, HE K M, 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. doi: 10.1109/TPAMI.2016.2577031
4	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2016: 779-788.
5	REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2017: 6517-6525.
6	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. [2023-09-17]. https://arxiv.org/abs/1804.02767.
7	BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. [2023-09-17]. http://arxiv.org/abs/2004.10934.
8	江波, 屈若锟, 李彦冬, 等. 基于深度学习的无人机航拍目标检测研究综述. 航空学报, 2021, 42(4): 524519. URL
	JIANG B, QU R K, LI Y D, et al. Object detection in UAV imagery based on deep learning: review. Acta Aeronautica et Astronautica Sinica, 2021, 42(4): 524519. URL
9	柯鹏飞, 蔡茂国, 吴涛. 基于改进卷积神经网络与集成学习的人脸识别算法. 计算机工程, 2020, 46(2): 262-267, 273. URL
	KE P F, CAI M G, WU T. Face recognition algorithm based on improved convolutional neural network and ensemble learning. Computer Engineering, 2020, 46(2): 262-267, 273. URL
10	HOWARD A G, ZHU M L, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[EB/OL]. [2023-09-17]. http://arxiv.org/abs/1704.04861.
11	SANDLER M, HOWARD A, ZHU M L, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 4510-4520.
12	HOWARD A, SANDLER M, CHEN B, et al. Searching for MobileNetV3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). Washington D. C., USA: IEEE Press, 2019: 1314-1324.
13	ZHANG X Y, ZHOU X Y, LIN M X, et al. ShuffleNet: an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 6848-6856.
14	MA N N, ZHANG X Y, ZHENG H T, et al. ShuffleNet V2: practical guidelines for efficient CNN architecture design[M]. Berlin, Germany: Springer International Publishing, 2018.
15	TAN M X, LE Q V. EfficientNet: rethinking model scaling for convolutional neural networks[EB/OL]. [2023-09-17]. http://arxiv.org/abs/1905.11946.
16	WANG C Y, MARK LIAO H Y, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Washington D. C., USA: IEEE Press, 2020: 1571-1580.
17	LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2017: 936-944.
18	LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 8759-8768.
19	HAN K, WANG Y H, TIAN Q, et al. GhostNet: more features from cheap operations[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2020: 1577-1586.
20	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2016: 770-778.
21	WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 3-19.
22	HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 7132-7141.
23	WANG Q L, WU B G, ZHU P F, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2020: 11531-11539.
24	ZHENG Z H, WANG P, LIU W, et al. Distance-IoU loss: faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2020: 12993-13000.
25	TONG Z J, CHEN Y H, XU Z W, et al. Wise-IoU: bounding box regression loss with dynamic focusing mechanism[EB/OL]. [2023-09-17]. http://arxiv.org/abs/2301.10051.
26	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proceedings of the European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 21-37.

[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]	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.
[10]	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.
[11]	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.
[12]	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.
[13]	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.
[14]	王炼红, 林飞鹏, 李潇瑶, 谌桂枝, 周莉. 融入课程知识图谱的KMAKT预测[J]. 计算机工程, 2024, 50(7): 23-31.
[15]	钱来, 赵卫伟. 基于对比学习和注意力机制的文本分类方法[J]. 计算机工程, 2024, 50(7): 104-111.

选择文件类型/文献管理软件名称

选择包含的内容