YOLO-LMC: An Enhanced Framework for Robust Object Detection
DOI:
https://doi.org/10.6919/ICJE.202607_12(7).0010Keywords:
Hazy Scenes; Object Detection; YOLO-LMC; YOLO-v8; Latent Space Metric Learning; Positive Sample Assignment.Abstract
Object detection in hazy scenes holds significant real-world implications, particularly in areas such as intelligent transportation and environmental monitoring. However, image quality degradation caused by haze inherently weakens the feature extraction capabilities of neural network models, leading to reduced accuracy. To address this challenge, this paper proposes YOLO-LMC (YOLO with Latent-space Metric Constraints), a novel architecture that integrates metric learning within the latent space optimization of object detection. Specifically, a positive sample assignment strategy is designed to filter high-confidence features within the latent space. This strategy, coupled with a cosine distance metric loss function based on class centers, effectively compels intra-class samples to form compact distributions in the latent space while ensuring distinct inter-class separability. Experimen-tal validation on the Foggy-Cityscapes dataset demonstrates that the proposed method significantly outperforms other object detection enhancement approaches, achieving an impressive at least 8.6% improvement in mAP.
Downloads
References
[1] Nazir, A., & Wani, M. A. (2023). You only look once-object detection models: A review. In Proceedings of the 10th International Conference on Computing for Sustainable Global Development (INDIACom) (pp. 1088–1095). IEEE.
[2] Ren, S., He, K., Girshick, R., & Sun, J. (2016). Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 1137–1149.
[3] Yang, S., Lu, H., & Li, J. (2022). Multifeature fusion-based object detection for intelligent transportation systems. IEEE Transactions on Intelligent Transportation Systems, 24, 1126–1133.
[4] Feng, D., Harakeh, A., Waslander, S. L., & Dietmayer, K. (2021). A review and comparative study on probabilistic object detection in autonomous driving. IEEE Transactions on Intelligent Transportation Systems, 23, 9961–9980.
[5] Abba, S., Bizi, A. M., Lee, J. A., Bakouri, S., & Crespo, M. L. (2024). Real-time object detection, tracking, and monitoring framework for security surveillance systems. Heliyon, 10.
[6] Gupta, H., Kotlyar, O., Andreasson, H., & Lilienthal, A. J. (2024). Robust object detection in challenging weather conditions. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (pp. 7523–7532).
[7] Li, C., Zhou, H., Liu, Y., Yang, C., Xie, Y., Li, Z., & Zhu, L. (2023). Detection-friendly dehazing: Object detection in real-world hazy scenes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45, 8284–8295.
[8] Mullappilly, S. S., Gehlot, A. S., Anwer, R. M., Khan, F. S., & Cholakkal, H. (2024). Semi-supervised open-world object detection. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 38, pp. 4305–4314).
[9] Husain, A. A., Maity, T., & Yadav, R. K. (2020). Vehicle detection in intelligent transport system under a hazy environment: A survey. IET Image Processing, 14, 1–10.
[10] Sharma, T., Nalla, B. T., Verma, N. K., & Vasikarla, S. (2022). FR-HDNet: Faster RCNN based haze detection network for image dehazing. In Proceedings of the 2022 IEEE Applied Imagery Pattern Recognition Workshop (AIPR) (pp. 1–8). IEEE.
[11] Cai, B., Xu, X., Jia, K., Qing, C., & Tao, D. (2016). Dehazenet: An end-to-end system for single image haze removal. IEEE Transactions on Image Processing, 25, 5187–5198.
[12] Zhu, X., Pang, J., Yang, C., Shi, J., & Lin, D. (2019). Adapting object detectors via selective cross-domain alignment. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 687–696).
[13] Zhao, Z., Guo, Y., & Ye, J. (2020). Bi-dimensional feature alignment for cross-domain object detection. In Proceedings of the European Conference on Computer Vision (pp. 671–686). Springer.
[14] Hnewa, M., & Radha, H. (2021). Multiscale domain adaptive YOLO for cross-domain object detection. In Proceedings of the 2021 IEEE International Conference on Image Processing (ICIP) (pp. 3323–3327). IEEE.
[15] Saito, K., Ushiku, Y., Harada, T., & Saenko, K. (2019). Strong-weak distribution alignment for adaptive object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 6956–6965).
[16] Song, W., Ren, S., Tan, W., & Wang, X. (2024). THNet: Transferability-aware hierarchical network for robust cross-domain object detection. IEEE Access.
[17] Varghese, R., & Sambath, M. (2024). YOLOv8: A novel object detection algorithm with enhanced performance and robustness. In Proceedings of the 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS) (pp. 1–6). IEEE.
[18] Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 2117–2125).
[19] van Doorenmalen, J., & Menkovski, V. (2020). Evaluation of CNN performance in semantically relevant latent spaces. In Proceedings of the International Symposium on Intelligent Data Analysis (pp. 145–157). Springer.
[20] Deng, L., Li, H., Liu, H., & Gu, J. (2022). A lightweight YOLOv3 algorithm used for safety helmet detection. Scientific Reports, 12, 10981.
[21] Feng, C., Zhong, Y., Gao, Y., Scott, M. R., & Huang, W. (2021). TOOD: Task-aligned one-stage object detection. In Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision (ICCV) (pp. 3490–3499). IEEE Computer Society.
[22] Wang, G., Li, S., Zhu, X., Wang, Y., Huang, J., Zhong, Y., & Wu, Z. (2025). REW-YOLO: A lightweight box detection method for logistics. Modelling, 6.
[23] Du, S., Zhang, B., Zhang, P., & Xiang, P. (2021). An improved bounding box regression loss function based on CIOU loss for multi-scale object detection. In Proceedings of the 2021 IEEE 2nd International Conference on Pattern Recognition and Machine Learning (PRML) (pp. 92–98). IEEE.
[24] Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., & Schiele, B. (2016). The cityscapes dataset for semantic urban scene understanding. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 3213–3223).
[25] Sakaridis, C., Dai, D., & Van Gool, L. (2018). Semantic foggy scene understanding with synthetic data. International Journal of Computer Vision, 126, 973–992.
[26] Joyce, J. M. (2011). Kullback-leibler divergence. In S. Kotz, C. B. Read, & N. L. Johnson (Eds.), International encyclopedia of statistical science (pp. 720–722). Springer.
[27] Cui, X., Zhang, W., Tüske, Z., & Picheny, M. (2018). Evolutionary stochastic gradient descent for optimization of deep neural networks. Advances in Neural Information Processing Systems, 31.
[28] Linderman, G. C., & Steinerberger, S. (2019). Clustering with t-SNE, provably. SIAM Journal on Mathematics of Data Science, 1, 313–332.
[29] Liu, W., Ren, G., Yu, R., Guo, S., Zhu, J., & Zhang, L. (2022). Image-adaptive YOLO for object detection in adverse weather conditions. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 36, pp. 1792–1800).
Downloads
Published
Issue
Section
License
Copyright (c) 2026 International Core Journal of Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
