对比特征增强的高架库小目标检测方法

doi:10.3778/j.issn.1002-8331.2307-0273

摘要/Abstract

摘要： 针对高架库区场景下安全帽检测中目标特征信息少、分类精度低等问题，提出小目标对比特征增强网络。首先提出快速空间金字塔池化跨层融合模块，减少空间维度上的目标信息丢失。然后提出小目标对比特征增强模块，使用双路并行空洞卷积获取不同感受野，利用通道注意力获取特征图在通道维度上更为精准的特征信息，采用浅层特征图减去深层特征图的方法削弱浅层特征图中大目标信息，以增强小目标特征信息表达。加入高效通道注意力解耦检测头，通过将检测头解耦为分类和回归分支，分别学习目标的语义信息和位置信息。实验结果表明，在TT100K数据集上，所提方法的mAP@0.5比基准网络YOLOv5提高了6.4个百分点，比YOLOv7提高了1.9个百分点。在自建高架库数据集上，所提方法的mAP@0.5相比基准网络提高了4.9个百分点，其中安全帽的mAP@0.5相比基准网络提高了6.9个百分点。

关键词: 小目标检测, 高架库, 跨层融合, 对比特征增强, 解耦检测头, YOLOv5

Abstract: In response to the issues of limited target feature information and low classification accuracy in safety helmet detection in elevated warehouse scenarios, a small target contrastive feature enhancement network is proposed. Firstly, a spatial pyramid pooling fast cross layer fusion module is introduced to reduce the loss of target information in the spatial dimension. Then, a small target contrastive feature enhancement module is presented, utilizing dual-path parallel dilated convolutions to capture different receptive fields, and incorporating channel attention to obtain more precise feature information in the channel dimension. Additionally, the large object information in shallow feature maps is weakened by subtracting them from deep feature maps, aiming to enhance the expression of small object features. Finally, an efficient channel attention decoupled detection head is incorporated, separating the detection head into classification and regression branches to learn semantic and positional information of the targets, respectively. Experimental results on the TT100K dataset demonstrate that the proposed method improves the mAP@0.5 compared to the YOLOv5 baseline network by 6.4 percentage points and outperforms YOLOv7 by 1.9 percentage points. Moreover, on a self-built elevated warehouse dataset, the method achieves a 4.9 percentage points improvement in mAP@0.5 compared to the baseline network, and a 6.9 percentage points increase in mAP@0.5 specifically for safety helmets.

Key words: small target detection, elevated warehouse, cross-layer fusion, contrastive feature enhancement, decoupled detection head, YOLOv5

朱贺, 卞长智, 张婧, 王力, 李小霞, 陈禹伶. 对比特征增强的高架库小目标检测方法[J]. 计算机工程与应用, 2024, 60(22): 347-354.

ZHU He, BIAN Changzhi, ZHANG Jing, WANG Li, LI Xiaoxia, CHEN Yuling. Contrastive Feature Enhancement for Elevated Warehouse Small Target Detection Method[J]. Computer Engineering and Applications, 2024, 60(22): 347-354.

参考文献

[1] 许凯, 邓超. 基于改进YOLOv3的安全帽佩戴识别算法[J]. 激光与光电子学进展, 2021, 58(6): 300-307.
XU K, DENG C. Research on helmet wear identification based on improved YOLOv3[J]. Laser & Optoelectronics Progress, 2021, 58(6): 300-307.
[2] KELM A, LAU?AT L, MEINS-BECKER A, et al. Mobile passive radio frequency identification (RFID) portal for auto-mated and rapid control of personal protective equipment (PPE) on construction sites[J]. Automation in Construction, 2013, 36: 38-52.
[3] BARRO-TORRES S, FERNáNDEZ-CARAMéS T M, PéREZ-IGLESIAS H J, et al. Real-time personal protective equipment monitoring system[J]. Computer Communications, 2012, 36(1): 42-50.
[4] RUBAIYAT A H M, TOMA T T, KALANTARI-KHANDANI M, et al. Automatic detection of helmet uses for construction safety[C]//Proceedings of the 2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops, 2016: 135-142.
[5] SHRESTHA K, SHRESTHA P P, BAJRACHARYA D, et al. Hard-hat detection for construction safety visualization[J]. Journal of Construction Engineering, 2015(1): 1-8.
[6] WARANUSAST R, BUNDON N, TIM TONG V, et al. Machine vision techniques for motorcycle safety helmet detection[C]//Proceedings of the 28th International Conference on Image and Vision Computing, New Zealand, 2013: 35-40.
[7] DOUNGMALA P, KLUBSUWAN K. Helmet wearing detection in Thailand using Haar like feature and circle Hough transform on image processing[C]//Proceedings of the 2016 IEEE International Conference on Computer and Information Technology, 2016: 611-614.
[8] JIA J, BAO Q, TANG H. Method for detecting safety helmet based on deformable part model[J]. Application Research of Computers, 2016, 33(3): 953-956.
[9] LI K, ZHAO X, BIAN J, et al. Automatic safety helmet wearing detection[C]//Proceedings of the 7th IEEE Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, Honolulu, 2017: 617-622.
[10] WU H, ZHAO J. Automated visual helmet identification based on deep convolutional neural networks[J]. Computer Aided Chemical Engineering, 2018, 44: 2299-2304.
[11] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision, 2015: 1440-1448.
[12] LONG X, CUI W, ZHENG Z. Safety helmet wearing detec-tion based on deep learning[C]//Proceedings of the 3rd IEEE Information Technology, Networking, Electronic and Automation Control Conference, 2019: 2495-2499.
[13] WU D M, WANG H, LI J. Safety helmet detection and identification based on improved faster R-CNN[J]. Information Technology & Informatization, 2020, 1: 17-20.
[14] XU X F, ZHAO W F, ZOU H Q, et al. Detection algorithm of safety helmet wear based on MobileNet-SSD[J]. Computer Engineering, 2021, 47: 298-305.
[15] DENG B Y, LEI X C, YE M. Safety helmet detection method based on YOLOv4[C]//Proceedings of the 16th IEEE International Conference on Computational Intelligence and Security, 2020: 155-158.
[16] SADIQ M, MASOOD S, PAL O. FD-YOLOv5: a fuzzy image enhancement based robust object detection model for safety helmet detection[J]. International Journal of Fuzzy Systems, 2022, 24(5): 2600-2616.
[17] BAI Y, ZHANG Y, DING M, et al. SOD-MTGAN: small object detection via multi-task generative adversarial network[C]//Proceedings of the 15th European Conference on Computer Vision, 2018: 206-221.
[18] LIN T Y, DOLLáR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
[19] DENG C, WANG M, LIU L, et al. Extended feature pyramid network for small object detection[J]. IEEE Transactions on Multimedia, 2021, 24: 1968-1979.
[20] MIN K, LEE G H, LEE S W. Attentional feature pyramid network for small object detection[J]. Neural Networks, 2022, 155: 439-450.
[21] CHEN S, ZHAO J, ZHOU Y, et al. Info-FPN: an informative feature pyramid network for object detection in remote sensing images[J]. Expert Systems with Applications, 2023, 214: 119-132.
[22] LUO Y, CAO X, ZHANG J, et al. CE-FPN: enhancing channel information for object detection[J]. Multimedia Tools and Applications, 2022, 81(21): 30685-30704.
[23] TAN M, PANG R, LE Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10781-10790.
[24] WANG Q, WU B, ZHU P, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]//Pro-ceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11534-11542.
[25] LIU S, QI L, QIN H, et al. Path aggregation network for instance segmentation[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8759-8768.
[26] GE Z, LIU S, WANG F, et al. YOLOX: exceeding YOLO series in 2021[J]. arXiv:2107.08430, 2021.
[27] LI C, LI L, JIANG H, et al. YOLOv6: a single-stage object detection framework for industrial applications[J]. arXiv:2209.02976, 2022.
[28] 孟繁星, 于瓅. 基于YOLOv5-EA的交通标志识别[J]. 辽宁工业大学学报 (自然科学版), 2022, 42(5): 303-310.
MENG F X, YU L. Traffic sign recognition based on YOLOv5-EA[J]. Journal of Liaoning University of Technology (Natural Science Edition), 2022, 42(5): 303-310.
[29] ZHU X, LYU S, WANG X, et al. TPH-YOLOv5: improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision, 2021: 2778-2788.
[30] WANG C Y, BOCHKOVSKIY A, LIAO H Y 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, 2023: 7464-7475.