基于ResNet50_SIMAM的水下目标检测模型研究
Research on Underwater Target Detection Model Based on ResNet50_SIMAM
DOI: 10.12677/CSA.2024.143058, PDF,    科研立项经费支持
作者: 柏填晟, 张亚婷, 金 珊, 李晓璇, 刘朝霞*：大连外国语大学软件学院，辽宁 大连
关键词: 水下目标检测；深度学习；ResNet50；TensorFlow；SimAM；Underwater Object Detection； Deep Learning； ResNet50； TensorFlow； SimAM
摘要: 水下目标检测是海洋探索和监测领域的一个关键技术挑战，具有广泛的应用。由于水下环境复杂以及视觉清晰度有限，现有检测方法效果不佳。针对这一问题，我们对现有的ResNet50模型进行改进，通过引入SIMAM注意力机制来提高检测精确度。通过对数据集的预处理和增强，模型成功适应了水下图像的特点。实验结果表明，该模型在水下目标检测任务上表现卓越，Map值由原来的64.6上升到68.35，验证了改进后的模型ResNet50_SIMAM在处理复杂水下视觉任务中的巨大潜力。
Abstract: Underwater target detection is a key technological challenge in the field of ocean exploration and monitoring and has a wide range of applications. Due to the complexity of the underwater environment as well as the limited visual clarity, the existing detection methods are ineffective. To address this problem, we improve the existing ResNet50 model to enhance the detection accuracy by intro-ducing the SIMAM attention mechanism. By preprocessing and enhancing the dataset, the model is successfully adapted to the characteristics of underwater images. The experimental results show that the model performs excellently on the underwater target detection task, and the Map value increases from the original 64.6 to 68.35, which verifies the great potential of the improved model ResNet50_SIMAM in dealing with the complex underwater vision tasks.
文章引用：柏填晟, 张亚婷, 金珊, 李晓璇, 刘朝霞. 基于ResNet50_SIMAM的水下目标检测模型研究[J]. 计算机科学与应用, 2024, 14(3): 58-65. https://doi.org/10.12677/CSA.2024.143058

参考文献

[1] Chen, L., Yin, T., Zhou, S., et al. (2023) Underwater Target Detection Algorithm Based on Feature Fusion Enhancement. Electronics, 12, Article 2756. [Google Scholar] [CrossRef] 
[2] Wang, X., Xue, G., Huang, S., et al. (2023) Underwater Object Detection Algorithm Based on Adding Channel and Spatial Fusion Attention Mechanism. Journal of Marine Science and Engineering, 11, Article 1116. [Google Scholar] [CrossRef] 
[3] Chen, R. and Chen, Y. (2023) Improved Convolutional Neural Net-work YOLOv5 for Underwater Target Detection Based on Autonomous Underwater Helicopter. Journal of Marine Sci-ence and Engineering, 11, Article 989. [Google Scholar] [CrossRef] 
[4] Hua, X., Cui, X.P., Xu, X.H., et al. (2023) Underwater Object Detec-tion Algorithm Based on Feature Enhancement and Progressive Dynamic Aggregation Strategy. Pattern Recognition, 139, Article ID: 109511. [Google Scholar] [CrossRef] 
[5] Song, P.H., Li, P.T., Dai, L.H., et al. (2023) Boosting R-CNN: Reweighting R-CNN Samples by RPN’s Error for Underwater Object Detection. Neurocomputing, 530, 150-164.
[6] Xu, S.B., Zhang, M.H., Song, W., et al. (2023) A Systematic Review and Analysis of Deep Learn-ing-Based Underwater Object Detection. Neurocomputing, 527, 204-232.
[7] Chen, L., Yang, Y.Y., Wang, Z.H., et al. (2023) Lightweight Underwater Target Detection Algorithm Based on Dynamic Sampling Transformer and Knowledge-Distillation Optimization. Journal of Marine Science and Engineering, 11, Article 426. [Google Scholar] [CrossRef] 
[8] Chen, Y., Song, P.H., Liu, H., et al. (2023) Achieving Domain Gener-alization for Underwater Object Detection by Domain Mixup and Contrastive Learning. Neurocomputing, 528, 20-34.
[9] Fu, C.P., Liu, R.S., Fan, X., et al. (2023) Rethinking General Underwater Object Detection: Datasets, Challenges, and Solutions. Neurocomputing, 517, 243-256.
[10] Yuan, W.Q., Fu, C.P., Liu, R.S., et al. (2022) SSoB: Searching a Scene-Oriented Architecture for Underwater Object Detection. The Visual Computer, 39, 5199-5208. [Google Scholar] [CrossRef] 
[11] Cheng, P.Y., Ting, Y.Z., Dong, G.L, et al. (2023) Water Source Identification in Mines Combining LIF Technology and ResNet. Journal of Mountain Science, 20, 3392-3401. [Google Scholar] [CrossRef] 
[12] Xu, L.Y., Huang, J.J., Zhang, H., et al. (2022) Direction of Arri-val Estimation of Acoustic Sources with Unmanned Underwater Vehicle Swarm via Matrix Completion. Remote Sensing, 14, Article 3790. [Google Scholar] [CrossRef] 
[13] Fayaz, S., Parah, S.A., et al. (2022) Underwater Object Detection: Archi-tectures and Algorithms – A Comprehensive Review. Multimedia Tools and Applications, 81, 20871-20916. [Google Scholar] [CrossRef] 
[14] Wang, X.H., Zhu, Y.G., Li, D.Y. et al. (2022) Underwater Tar-get Detection Based on Reinforcement Learning and Ant Colony Optimization. Journal of Ocean University of China, 21, 323-330. [Google Scholar] [CrossRef] 
[15] Qi, J.H., Gong, Z.Q., Yao, A.H., et al. (2021) Bathy-metric-Based Band Selection Method for Hyperspectral Underwater Target Detection. Remote Sensing, 13, Article 3798. [Google Scholar] [CrossRef]