基于表观特征增强及融合的安全帽检测算法

doi:10.3778/j.issn.1002-8331.2404-0465

摘要/Abstract

摘要： 佩戴安全帽是保障工人生命安全的重要措施之一。针对安全帽佩戴检测网络难以准确定位和识别与环境颜色相似的安全帽和被遮挡的安全帽问题，提出一种基于表观特征增强及融合的安全帽检测算法。设计表观特征增强模块（appearance feature enhancement module，AFEM）来构建主干特征提取网络。该模块通过并行使用浅层分支、倒置瓶颈结构分支和非对称卷积分支，实现了梯度向不同分支的传播，进而捕捉更多丰富的形状、边缘和纹理等外观特征；此外，提出双尺度特征融合模块（dual-scale feature fusion，DSFF）。该模块通过融合更细粒度的浅层特征图和特征金字塔中深层特征图来获取和强化被遮挡目标的局部表观特征信息，从而提高残缺目标的检测精度。实验结果表明，在SHWD数据集上，改进后的模型在保持实时性同时，召回率和精度分别达到88.3%和92.6%，较原模型提高了0.8和1.2个百分点。在检测不同场景下的安全帽时表现更为出色，能够更好地适应各种复杂多样场景。

关键词: 安全帽检测, YOLOv8, 特征提取, 特征融合

Abstract: Wearing safety helmets is one of the important measures to ensure the life and safety of workers. Aiming at the problem that the safety helmet wearing detection network is difficult to accurately locate and identify the safety helmet with similar color to the environment and the occluded safety helmet, this paper proposes a safety helmet detection algorithm based on appearance feature enhancement and fusion. The appearance feature enhancement module （AFEM) is designed to construct the backbone feature extraction network. By using shallow branches, inverted bottleneck structure branches and asymmetric convolution branches in parallel, the module realizes the propagation of gradient to different branches, thereby capturing more rich appearance features such as shape, edge and texture. In addition, a dual-scale feature fusion module （DSFF) is proposed. The module obtains and strengthens the local appearance feature information of the occluded target by fusing the finer-grained shallow feature map and the deep feature map in the feature pyramid, so as to improve the detection accuracy of the incomplete target. Experimental results show that on the SHWD dataset, the improved model achieves 88.3%recall and 92.6% precision while maintaining real-time performance, which is 0.8 and 1.2 percentage points higher than the original model. It performs better in detecting safety helmets in different scenes, and can better adapt to various complex and diverse scenes.

Key words: safety helmet detection, YOLOv8, feature extraction, feature fusion

杨国亮, 洪鑫芳, 盛杨杨, 熊文楷. 基于表观特征增强及融合的安全帽检测算法[J]. 计算机工程与应用, 2025, 61(15): 363-372.

YANG Guoliang, HONG Xinfang, SHENG Yangyang, XIONG Wenkai. Helmet Detection Algorithm Based on Appearance Feature Enhancement and Fusion[J]. Computer Engineering and Applications, 2025, 61(15): 363-372.

参考文献

[1] 张立艺, 武文红, 牛恒茂, 等. 深度学习中的安全帽检测算法应用研究综述[J]. 计算机工程与应用, 2022, 58(16): 1-17.
ZHANG L Y, WU W H, NIU H M, et al. Summary of application research on helmet detection algorithm based on deep learning[J]. Computer Engineering and Applications, 2022, 58(16): 1-17.
[2] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2015: 1440-1448.
[3] 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.
[4] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//Proceedings of the 14th European Conference on Computer Vision. Cham: Springer International Publishing, 2016: 21-37.
[5] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 6517-6525.
[6] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[J]. arXiv:2004. 10934, 2020.
[7] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 779-788.
[8] HE K M, GKIOXARI G, DOLLáR P, et al. Mask R-CNN[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 2980-2988.
[9] 刘晓慧, 叶西宁. 肤色检测和Hu矩在安全帽识别中的应用[J]. 华东理工大学学报(自然科学版), 2014, 40(3): 365-370.
LIU X H, YE X N. Skin color detection and Hu moments in helmet recognition research[J]. Journal of East China University of Science and Technology (Natural Science Edition), 2014, 40(3): 365-370.
[10] 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. Piscataway: IEEE, 2016: 135-142.
[11] 赵红成, 田秀霞, 杨泽森, 等. YOLO-S: 一种新型轻量的安全帽佩戴检测模型[J]. 华东师范大学学报(自然科学版), 2021(5): 134-145.
ZHAO H C, TIAN X X, YANG Z S, et al. YOLO-S: a new lightweight helmet wearing detection model[J]. Journal of East China Normal University (Natural Science), 2021(5): 134-145.
[12] 宋晓凤, 吴云军, 刘冰冰, 等. 改进YOLOv5s算法的安全帽佩戴检测[J]. 计算机工程与应用, 2023, 59(2): 194-201.
SONG X F, WU Y J, LIU B B, et al. Improved YOLOv5s algorithm for helmet wearing detection[J]. Computer Engineering and Applications, 2023, 59(2): 194-201.
[13] 李嘉信, 胡杨, 黄协舟, 等. 面向小目标的多空间层次安全帽检测[J]. 计算机工程与应用, 2024, 60(6): 230-237.
LI J X, HU Y, HUANG X Z, et al. Small target-oriented multi-space hierarchical helmet detection[J]. Computer Engineering and Applications, 2024, 60(6): 230-237.
[14] 吕宗喆, 徐慧, 杨骁, 等. 面向小目标的YOLOv5安全帽检测算法[J]. 计算机应用, 2023, 43(6): 1943-1949.
LYU Z Z, XU H, YANG X, et al. Small object detection algorithm of YOLOv5 for safety helmet[J]. Journal of Computer Applications, 2023, 43(6): 1943-1949.
[15] LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 8759-8768.
[16] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 770-778.
[17] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 2261-2269.
[18] HOWARD A, SANDLER M, CHEN B, et al. Searching for MobileNetV3[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2019: 1314-1324.
[19] MA N N, ZHANG X Y, ZHENG H T, et al. ShuffleNet V2: practical guidelines for efficient CNN architecture design[C]//Proceedings of the European Conference on Computer Vision. Cham: Springer, 2018: 122-138.
[20] HAN K, WANG Y H, TIAN Q, et al. GhostNet: more features from cheap operations[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 1577-1586.
[21] 黄硕清, 黄金贵. 基于RFB和YOLOv5特征增强融合改进的钢材表面缺陷检测方法[J]. 计算机工程, 2025, 51(4): 249-260.
HUANG S Q, HUANG J G. Improved steel defect detection method based on enhanced fusion of RFB and YOLOv5 features[J]. Computer Engineering, 2025, 51(4): 249-260.
[22] 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.
[23] LEE Y, HWANG J W, LEE S, et al. An energy and GPU-computation efficient backbone network for real-time object detection[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway: IEEE, 2019: 752-760.
[24] JIN Z C, YU D D, SONG L C, et al. You should look at all objects[C]//Proceedings of the European Conference on Computer Vision. Cham: Springer Nature Switzerland, 2022: 332-349.
[25] TAN M X, PANG R M, LE Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 10778-10787.
[26] LUO Y H, CAO X, ZHANG J T, et al. CE-FPN: enhancing channel information for object detection[J]. Multimedia Tools and Applications, 2022, 81(21): 30685-30704.
[27] LIU S, HUANG D, WANG Y. Learning spatial fusion for single-shot object detection[J]. arXiv:1911.09516, 2019.
[28] 李雅雯, 孙浩然, 胡跃明, 等. 基于注意力机制与多尺度特征融合的电极缺陷YOLO检测算法[J]. 控制与决策, 2023, 38(9): 2578-2586.
LI Y W, SUN H R, HU Y M, et al. Electrode defect YOLO detection algorithm based on attention mechanism and multi-scale feature fusion[J]. Control and Decision, 2023, 38(9): 2578-2586.
[29] 苗荣慧, 李港澳, 黄宗宝, 等. 基于YOLOv7-ST-ASFF的复杂果园环境下苹果成熟度检测方法[J]. 农业机械学报, 2024, 55(6): 219-228.
MIAO R H, LI G A, HUANG Z B, et al. Maturity detection of apple in complex orchard environment based on YOLOv7-ST-ASFF[J]. Transactions of the Chinese Society for Agricultural Machinery, 2024, 55(6): 219-228.