Abstract
Numerous deaths are being reported at sites where safety helmets are important for workers. Though these sites are usually equipped with closed circuit television (CCTV) cameras, real-time monitoring of these feeds proves to be a mammoth task as it requires large human personnel leading to financial burdens. This work harnesses the object detection abilities of various deep learning models to propose a robust solution for safety helmet detection problem. This work divides the overall task into three sub problems, where the first two sub problems explore various object detection models including Single Shot Detector (SSD), Faster Region Based Convolutional Neural Networks (FRCNN), YOLOv3, YOLOv4, and YOLOv5 over multiple benchmark datasets. In Sub Problem 1, the YOLOv5 model achieved a mAP of 93.5%, 94.1% and 92.7% on the three selected datasets, while it achieved a mAP of 93.8% over the selected dataset of the Sub Problem 2. In both these sub problems, the YOLOv5 model outperformed all the recently proposed models for helmet detection task and hence was selected as the model for further experiments. In the final sub problem, a novel fusion of YOLOv5 and a fuzzy based image enhancement module, was further fine-tuned to create a robust model (FD-YOLOv5 M) that worked visibly better than simple YOLOv5 even over real site noisy CCTV feeds. The proposed FD-YOLOv5 model, when tried over an enhanced dataset, efficiently detected safety helmets worn by humans and even differentiated them from various types of other common headgears. These observations strongly support the application of the proposed model at real site scenarios to detect workers wearing safety helmets.

















Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.References
Fang, Q., Li, H., Luo, X., et al.: Detecting non-hard hat-use by a deep learning method from far-field surveillance videos. Autom. Constr. 85, 1–9 (2018). https://doi.org/10.1016/j.autcon.2017.09.018
Fuchuan, RongXin, W.: Research on safety helmet wearing Yolo-V3 detection technology improvement in Mine Environment. J. Phys. Conf. Ser. 1345, 042045 (2019). https://doi.org/10.1088/1742-6596/1345/4/042045
Benyang, D., Xiaochun, L., Miao, Y.: Safety helmet detection method based on YOLOv4. In: 2020 16th International Conference on Computational Intelligence and Security (CIS) (2020). https://doi.org/10.1109/cis52066.2020.00041
Li, Y., Wei, H., Han, Z., et al.: Deep learning-based safety helmet detection in engineering management based on convolutional neural networks. Adv. Civ. Eng. 2020, 1–10 (2020). https://doi.org/10.1155/2020/9703560
Wang, H., Hu, Z., Guo, Y., et al.: A real-time safety helmet wearing detection approach based on CS YOLOv3. Appl. Sci. 10, 6732 (2020). https://doi.org/10.3390/app10196732
Njvisionpower. “NJVISIONPOWER/Safety-helmet-wearing-dataset: safety helmet wearing detect dataset, with pretrained model.” GitHub. https://github.com/njvisionpower/Safety-Helmet-Wearing-Dataset. Accessed 18 Feb 2022
China, Northeastern University -. “Hard hat workers object detection dataset.” Roboflow. https://public.roboflow.com/object-detection/hard-hat-workers. Accessed 30 Dec 2020
K&K Technologies, Inc. “Hard hat workers dataset: makeml - create neural network with ease.” MakeML. https://makeml.app/datasets/hard-hat-workers. Accessed 18 Feb 2022
Yao, S., Lin, W., Ong, E., Lu, Z.: Contrast signal-to-noise ratio for image quality assessment. In: IEEE International Conference on Image Processing, p. I-397 (2005). https://doi.org/10.1109/ICIP.2005.1529771
Liu, W., Anguelov, D., Erhan, D., et al.: SSD: single shot multibox detector. In: Computer Vision—ECCV 2016, pp. 21–37 (2016). https://doi.org/10.1007/978-3-319-46448-0_2
Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015). https://arxiv.org/abs/1506.01497
Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You Only Look Once: Unified, Real-Time Object Detection (2015). https://arxiv.org/abs/1506.02640
Redmon, J., Farhadi, A.: YOLOv3: An Incremental Improvement (2018). https://arxiv.org/abs/1804.02767
Bochkovskiy, A., Wang, C., Liao, H.: YOLOv4: Optimal Speed and Accuracy of Object Detection (2020). arXiv.org:2004.10934
Jocher, G., Stoken, A., Borovec, J.: ultralytics/yolov5: v5.0—YOLOv5-P6 1280 models, AWS, Supervise.ly and YouTube integrations (2021)
Hung, P., Kien, N.: SSD-MobileNet implementation for classifying fish species. In: Advances in Intelligent Systems and Computing, pp. 399–408 (2019). https://doi.org/10.1007/978-3-030-33585-4_40
Biswas, D., Su, H., Wang, C., et al.: An automatic traffic density estimation using Single Shot Detection (SSD) and MobileNet-SSD. Phys. Chem. Earth A/B/C 110, 176–184 (2019). https://doi.org/10.1016/j.pce.2018.12.001
Hu, X., Li, H., Li, X., Wang, C.: MobileNet-SSD Microscope using adaptive error correction algorithm: real-time detection of license plates on mobile devices. IET Intell. Transp. Syst. 14, 110–118 (2020). https://doi.org/10.1049/iet-its.2019.0380
Benjdira, B., Khursheed, T., Koubaa, A., et al.: Car detection using unmanned aerial vehicles: comparison between faster R-CNN and YOLOv3. In: 2019 1st International Conference on Unmanned Vehicle Systems (UVS), Oman (2019). https://doi.org/10.1109/uvs.2019.8658300
Tian, Y., Yang, G., Wang, Z., et al.: Apple detection during different growth stages in orchards using the improved YOLO-v3 model. Comput. Electron. Agric. 157, 417–426 (2019). https://doi.org/10.1016/j.compag.2019.01.012
Yi, Z., Yongliang, S., Jun, Z.: An improved tiny-YOLOv3 pedestrian detection algorithm. Optik 183, 17–23 (2019). https://doi.org/10.1016/j.ijleo.2019.02.038
Zhao, L., Li, S.: Object detection algorithm based on improved YOLOv3. Electronics 9, 537 (2020). https://doi.org/10.3390/electronics9030537
Li, C., Wang, R., Li, J., Fei, L.: Face detection based on YOLOv3. In: Recent Trends in Intelligent Computing. Communication and Devices, pp. 277–284 (2019). https://doi.org/10.1007/978-981-13-9406-5_34
Kuznetsova, A., Maleva, T., Soloviev, V.: Detecting apples in orchards using YOLOv3 and YOLOv5 in general and close-up images. In: Advances in Neural Networks—ISNN 2020, pp. 233–243 (2020). https://doi.org/10.1007/978-3-030-64221-1_20
Yang, G., Feng, W., Jin, J., et al.: Face mask recognition system with YOLOv5 based on image recognition. In: 2020 IEEE 6th International Conference on Computer and Communications (ICCC) (2020). https://doi.org/10.1109/iccc51575.2020.9345042
Kasper-Eulaers, M., Hahn, N., Berger, S., et al.: Short Communication: detecting heavy goods vehicles in rest areas in winter conditions using YOLOv5. Algorithms 14, 114 (2021). https://doi.org/10.3390/a14040114
Zhu, Q., Zheng, H., Wang, Y., et al.: Study on the evaluation method of sound phase cloud maps based on an improved YOLOv4 algorithm. Sensors 20, 4314 (2020). https://doi.org/10.3390/s20154314
Yu, Z., Shen, Y., Shen, C.: A real-time detection approach for bridge cracks based on YOLOv4-FPM. Autom. Constr. 122, 103514 (2021). https://doi.org/10.1016/j.autcon.2020.103514
Yu, J., Zhang, W.: Face mask wearing detection algorithm based on improved YOLO-v4. Sensors 21, 3263 (2021). https://doi.org/10.3390/s21093263
Albahli, S., Nida, N., Irtaza, A., et al.: Melanoma lesion detection and segmentation using YOLOv4-DarkNet and active contour. IEEE Access 8, 198403–198414 (2020). https://doi.org/10.1109/access.2020.3035345
Wu, D., Lv, S., Jiang, M.: Using channel pruning-based YOLOv4 deep learning algorithm for the real-time and accurate detection of apple flowers in natural environments. Comput. Electron. Agric. 178, 105742 (2020). https://doi.org/10.1016/j.compag.2020.105742
NVlabs. “NVlabs/FFHQ-dataset: flickr-faces-hq dataset (FFHQ).” GitHub. https://github.com/NVlabs/ffhq-dataset. Accessed 18 Feb 2022
K&K Technologies, Inc. “Helmets dataset: makeml create neural network with ease.” MakeML. https://makeml.app/datasets/helmets. Accessed 18 Feb 2022
Morillas, S., Gregori, V., Hervas, A.: Fuzzy peer groups for reducing mixed Gaussian-impulse noise from color images. IEEE Trans. Image Process. 18, 1452–1466 (2009). https://doi.org/10.1109/tip.2009.2019305
Acknowledgements
We would like to thank S. P. Singla Constructions Pvt. Ltd. for providing real time images and videos to validate this model on. Without their cooperation, certain aspects of this study would have remained unattended.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sadiq, M., Masood, S. & Pal, O. FD-YOLOv5: A Fuzzy Image Enhancement Based Robust Object Detection Model for Safety Helmet Detection. Int. J. Fuzzy Syst. 24, 2600–2616 (2022). https://doi.org/10.1007/s40815-022-01267-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40815-022-01267-2