Validation and Performance Contrast of Deep Neural Network Based Mechanism for Real-Time Automatic Safety Helmet Detection

International Journal of Electrical and Electronics Engineering

© 2024 by SSRG - IJEEE Journal

Volume 11 Issue 5

Year of Publication : 2024

Authors : Dasari Naga Vinod, N. Kapileswar, Judy Simon, Phani Kumar Polasi, B. Padmavathi, Partho Adhikari

10.14445/23488379/IJEEE-V11I5P110

How to Cite?

Dasari Naga Vinod, N. Kapileswar, Judy Simon, Phani Kumar Polasi, B. Padmavathi, Partho Adhikari, "Validation and Performance Contrast of Deep Neural Network Based Mechanism for Real-Time Automatic Safety Helmet Detection," SSRG International Journal of Electrical and Electronics Engineering, vol. 11,  no. 5, pp. 102-118, 2024. Crossref, https://doi.org/10.14445/23488379/IJEEE-V11I5P110

Abstract:

In the fourth industrial revolution, more people were working in production sites, including traffic areas, chemical plants, nuclear reactors, and building sites, raising worries about worker safety. The Head is the most important part of the body, and there have not been many studies done on detecting helmet use across a variety of units in hazardous environments. This research constructed a revolutionary deep learning mechanism called YOLOv8.0 that integrates object detection, key point localization, and basic rule-based reasoning to solve this problem. Moreover, it presents an active helmet-wearing detection method and a dataset created from scratch for multifunctional use applications. The three research questions that guide the process are, (i) Is it possible to identify certain classes in any video?, (ii) Can the model be used to identify helmets across numerous sites, and (iii) Can detections be made in challenging environmental circumstances in real-time? first created a dataset in Yolo format and augmented the photos to produce a second, more generic dataset. Next, the proposed datasets on 5 different versions of the YOLOv8.0 pre-trained model because YOLOv8.0 is an anchor-free mechanism; it detects object’s centers straightly rather than its redeem through a known anchor box. On the augmented dataset original dataset, the YOLOv8l version received the best mAP score among the other versions, scoring approximately 95% and 81% respectively.

Keywords:

 Computer vision, Image augmentation, Object detection, Safety helmet, SOTA DNN, YOLOV8.

References:

[1] Global Status Report on Road Safety 2018, World Health Organization, 2018.
[Google Scholar] [Publisher Link]
[2] WHO Sri Lanka, World Health Organization. [Online]. Available: https://www.who.int/srilanka
[3] Qi Wang, Junyu Gao, and Xuelong Li, “Weakly Supervised Adversarial Domain Adaptation for Semantic Segmentation in Urban Scenes,” IEEE Transactions on Image Processing, vol. 28, no. 9, pp. 4376-4386, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Shaoqing Ren et al., “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Kaiming He et al., “Mask R-CNN,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 2961-2969, 2017.
[Google Scholar] [Publisher Link]
[6] Joseph Redmon et al., “You Only Look Once: Unified, Real-Time Object Detection,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 779-788, 2016.
[Google Scholar] [Publisher Link]
[7] Wei Liu et al., “SSD: Single Shot Multibox Detector,” 14th European Conference, The Netherlands, pp. 21-37, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Tsung-Yi Lin et al., “Focal Loss for Dense Object Detection,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 2980-2988, 2017.
[Google Scholar] [Publisher Link]
[9] Shifeng Zhang et al., “Bridging the Gap between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 9759-976, 2020.
[Google Scholar] [Publisher Link]
[10] Zhi Tian et al., “FCOS: Fully Convolutional One-Stage Object Detection,” 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Korea (South), pp. 9626-9635, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Ze Yang et al., “Reppoints: Point Set Representation for Object Detection,” Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 9657-9666, 2019.
[Google Scholar] [Publisher Link]
[12] ultralytics/yolov5, Github. [Online]. Available: https://github.com/ultralytics/yolov5
[13] Linu Shine, and C.V. Jiji, “Automated Detection of Helmet on Motorcyclists from Traffic Surveillance Videos: A Comparative Analysis Using Hand-Crafted Features and CNN,” Multimedia Tools and Applications, vol. 79, pp. 14179-14199, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[14] C. Vishnu et al., “Detection of Motorcyclists without Helmet in Videos Using Convolutional Neural Network,” 2017 International Joint Conference on Neural Networks (IJCNN), USA, pp. 3036-3041, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[15] B. Yogameena, K. Menaka, and S. Saravana Perumaal, “Deep Learning‐Based Helmet Wear Analysis of a Motorcycle Rider for an Intelligent Surveillance System,” IET Intelligent Transport Systems, vol. 13, no. 7, pp. 1190-1198, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Lokesh Allamki et al., “Helmet Detection Using Machine Learning and Automatic License Plate Recognition,” International Research Journal of Engineering and Technology (IRJET), vol. 6, no. 12, pp. 80-84, 2019.
[Google Scholar] [Publisher Link]
[17] Fahad A. Khan, Nitin Nagori, and Ameya Naik, “Helmet and Number Plate detection of Motorcyclists Using Deep Learning and Advanced Machine Vision Techniques,” 2020 Second International Conference on Inventive Research in Computing Applications (ICIRCA), India, pp. 714-717, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[18] A. Farhadi, and J.  Redmon, “Yolov3: An Incremental Improvement,” Arxiv, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Apoorva Saumya et al., “Machine Learning Based Surveillance System for Detection of Bike Riders without Helmet and Triple Rides,” 2020 International Conference on Smart Electronics and Communication (ICOSEC), India, pp. 347-352, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[20] C.A. Rohith et al., “An Efficient Helmet Detection for MVD Using Deep Learning,” 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), India, pp. 282-286, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Aphinya Chairat et al., “Low Cost, High Performance Automatic Motorcycle Helmet Violation Detection,” Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 3560-3568, 2020.
[Google Scholar] [Publisher Link]
[22] Madhuchhanda Dasgupta, Oishila Bandyopadhyay, and Sanjay Chatterji, “Automated Helmet Detection for Multiple Motorcycle Riders Using CNN,” 2019 IEEE Conference on Information and Communication Technology, India, pp. 1-4, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Hanhe Lin et al., “Helmet Use Detection of Tracked Motorcycles Using CNN-Based Multi-Task Learning,” IEEE Access, vol. 8, pp. 162073-162084, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Felix Wilhelm Siebert, and Hanhe Lin, “Detecting Motorcycle Helmet Use with Deep Learning,” Accident Analysis and Prevention, vol. 134, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Wei Jia et al., “Real‐Time Automatic Helmet Detection of Motorcyclists in Urban Traffic Using Improved YOLOv5 Detector,” IET Image Processing, vol. 15, no. 14, pp. 3623-3637, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Munkh-Erdene Otgonbold et al., “SHEL5K: An Extended Dataset and Benchmarking for Safety Helmet Detection,” Sensors, vol. 22, no. 6, pp. 1-23, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Chuyi Li et al., “YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications,” ArXiv, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Chien-Yao Wang, Alexey Bochkovskiy, and Hong-Yuan Mark Liao, “YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 74647475, 2023.
[Google Scholar] [Publisher Link]
[29] Yiduo Zhang, Yi Qiu, and Huihui Bai, “FEFD-YOLOV5: A Helmet Detection Algorithm Combined with Feature Enhancement and Feature Denoising,” Electronics, vol. 12, no. 13, pp. 1-14, 2023.
[CrossRef] [Google Scholar] [Publisher Link]