A robust ensemble framework for helmet usage classification in real-world scenarios
and
*
Article Sidebar
Full Text: 
PDF
Main Article Content
Abstract
The application of machine learning models in the analysis of helmet-related images has yielded remarkable results in identifying and classifying helmet wearing behaviors. Previous research has used several pretrained models to predict proper or improper helmet use, achieving the highest accuracy of 98.61% on the Helmet Wearing Image Dataset (2024), a newly introduced dataset designed to improve the ability to classify helmet wearing behaviors. This study aims to improve the prediction performance on helmet datasets by leveraging state-of-the-art deep learning models combined with ensemble techniques. Using ResNet-50, MobileNetV2, and EfficientNet-B0 models, the proposed EnsemHelmet Framework uses soft voting ensemble to optimize the classification results, achieving an outstanding accuracy of 99.18% on the experimental dataset. The results demonstrate the potential of ensemble learning to achieve high performance. This study not only improves the accuracy of the helmet wearing recognition system but also highlights the effectiveness of ensemble techniques in optimizing performance on real-world datasets.
Article Details
Conflict of Interest
© 2026 The authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License.
References
Adewopo, V. A., Elsayed, N., ElSayed, Z., Ozer, M., Abdelgawad, A., & Bayoumi, M. (2023). A review on action recognition for accident detection in smart city transportation systems. Journal of Electrical Systems and Information Technology, 10(1), 57. https://doi.org/10.1186/s43067-023-00124-y
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 770–778). IEEE. https://doi.org/10.1109/CVPR.2016.90
Irfan, E., Jacob, C., & Resmi, R. (2024, May). Facial Recognition and CCTV Integration for Enhanced Security Using Deep Learning Techniques. In 2024 IEEE Recent Advances in Intelligent Computational Systems (RAICS) (pp. 1-5). IEEE. https://doi.org/10.1109/RAICS61201.2024.10689986
Masand, A., Chauhan, S., Jangid, M., Kumar, R., & Roy, S. (2021). Scrapnet: An efficient approach to trash classification. IEEE Access, 9, 130947–130958. https://doi.org/10.1109/ACCESS.2021.3111230
Nguyen, V. H., Bui, H. H. N., & Le, T. P. (2024, November). Assessing grain size variation across rice panicles using YOLOv8 and DeepLabv3 models. In Thai-Nghe, N., Do, T.N., & Benferhat, S. (Eds) Intelligent Systems and Data Science. ISDS 2024 (pp. 15–29). Springer. https://doi.org/10.1007/978-981-97-9616-8_2
Patil, K., Jadhav, R., Suryawanshi, Y., Chumchu, P., Khare, G., & Shinde, T. (2024). HelmetML: A dataset of helmet images for machine learning applications. Data in Brief, 56, 110790. https://doi.org/10.1016/j.dib.2024.110790
Ren, Y., Cong, L. (2023). Traffic Image Classification Algorithm Based on Deep-Learning. In Atiquzzaman, M., Yen, N., & Xu, Z. (Eds) Proceedings of the 4th International Conference on Big Data Analytics for Cyber-Physical System in Smart City - Volume 1. BDCPS 2022 (pp. 437–445). Springer. https://doi.org/10.1007/978-981-99-0880-6_48
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L. C. (2018). MobileNetV2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 4510–4520). IEEE. https://doi.org/10.1109/CVPR.2018.00474
Shourie, P. (2024). Optimizing traffic sign detection with MobileNetV2: A lightweight deep learning approach. In Proceedings of 2024 4th International Conference on Technological Advancements in Computational Sciences (ICTACS) (pp. 272–276). IEEE. https://doi.org/10.1109/ICTACS62700.2024.10840818
Singh, K., Patil, N., Mohite, S. G., Jadhav, S., Mohite, S., Gayakwad, M., & Joshi, R. (2024). Vehicle identification using RESNET-50: CNN approach. In Proceedings of 2024 IEEE International Conference on Blockchain and Distributed Systems Security (ICBDS) (pp. 1–6). IEEE. https://doi.org/10.1109/ICBDS61829.2024.10837029
Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (ICML), (pp. 6105–6114). PMLR.
Tran, N., Nguyen, H., Ly, D., & Nguyen, H. D. (2024, November). Violence detection using skeleton data with graph convolutional networks. In Thai-Nghe, N., Do, T.N., & Benferhat, S. (Eds) Intelligent Systems and Data Science. ISDS 2024 (pp. 86–97). Springer. https://doi.org/10.1007/978-981-97-9616-8_7
Truong, T. D., Huynh, P. H., Nguyen, V. H., & Do, T. N. (2024). Enhancing the efficiency of lung disease classification based on multi-modal fusion model. In Thai-Nghe, N., Do, TN., & Benferhat, S. (Eds) Intelligent Systems and Data Science. ISDS 2024 (pp. 86–97). Springer. https://doi.org/10.1007/978-981-97-9616-8_5
Vo, A. H., Son, L. H., Vo, M. T., & Le, T. (2019). A novel framework for trash classification using deep transfer learning. IEEE Access, 7, 178631–178639. https://doi.org/10.1109/ACCESS.2019.2959033
Vo, M. T., Vo, A. H., & Le, T. (2022). A robust framework for shoulder implant X-ray image classification. Data Technologies and Applications, 56(3), 447–460. https://doi.org/10.1108/DTA-08-2021-0210
Zhang, W., Dang, L. M., Nguyen, L. Q., Alam, N., Bui, N. D., Park, H. Y., & Moon, H. (2024). Adapting the Segment Anything Model for plant recognition and automated phenotypic parameter measurement. Horticulturae, 10(4), 398. https://doi.org/10.3390/horticulturae10040398