Robust Multi Scale Multi Polarization and Multi Orientation Fused Maritime Target Recognition in SAR Images

International Journal of Electronics and Communication Engineering

© 2025 by SSRG - IJECE Journal

Volume 12 Issue 6

Year of Publication : 2025

Authors : M Mary Rosaline, S Arivazhagan, Santhana Raj, Phillip Livingston, Stewart Phillip, K Alaguraja

10.14445/23488549/IJECE-V12I6P107

How to Cite?

M Mary Rosaline, S Arivazhagan, Santhana Raj, Phillip Livingston, Stewart Phillip, K Alaguraja, "Robust Multi Scale Multi Polarization and Multi Orientation Fused Maritime Target Recognition in SAR Images," SSRG International Journal of Electronics and Communication Engineering, vol. 12,  no. 6, pp. 90-100, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I6P107

Abstract:

A robust Multi Scale Multi Polarization and Multi Orientation CNN are proposed for feature extraction and classification in SAR-based Ship recognition, which exploits feature fusion across multiple scales, polarizations and looks. Data augmentation techniques are incorporated to mitigate data imbalance and increase the robustness of the model. The ablation studies are performed for 10 class classification problems of SAR ship classification varying the number of scales, orientations and polarizations with different learning algorithms and activation functions. The proposed model is validated on the OpenSARShip dataset to classify ships in SAR images. The proposed model achieved the highest accuracy of 97.1% for the task of ship classification. The proposed CNN model is better than the state-of-the-art conventional methods in terms of accuracy. The proposed model attained the said accuracy with only 328M network parameters. The proposed CNN is a beneficial model for identifying the different types of ships in SAR images and assisting maritime surveillance. The comparison of the experimental results with pre-trained and custom deep learning models available in the literature validates the reliability of the proposed deep CNN model.

Keywords:

Multi scale multi polarization and multi orientation CNN, Ship classification, Synthetic Aperture Radar.

References:

[1] Zhongzhen Sun et al., “Ship Recognition for Complex SAR Images via Dual-Branch Transformer Fusion Network,” IEEE Geoscience and Remote Sensing Letters, vol. 21, pp. 1-5, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[2] T. Saleh et al., “Ship Detection in COSMO-SkyMed SAR Imagery Using A Novel CNN-based Detector: A Case Study from the Suez Canal,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 10, pp. 715-722, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[3] O. Jean-Philippe et al., “Analysis of SAR Images by Multidimensional Wavelet Transform,” IEE proceedings of Radar, Sonar and Navigation, vol. 150, no. 4, pp. 234-241, 2003.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Huan Ruohong, and Yang Ruliang, “SAR Target Recognition Based on MRF and Gabor Wavelet Feature Extraction,” Proceedings of IEEE International Geoscience and Remote Sensing Symposium, Boston, MA, USA, pp. 907-910, 2008.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Wang Lu et al., “A Method of SAR Target Recognition Based on Gabor Filter and Local Texture Feature Extraction,” Journal of Radars, vol. 4, no. 6, pp. 658-665, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Yu Qiao, Yue Zhao, and Xiaoyong Men, “Target Recognition in SAR Images via Graph Wavelet Transform and 2DPCA,” Proceedings of the 2nd International Conference on Image and Graphics Processing, Singapore Singapore, pp. 3-7, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[7] R. Bhalla, and Hao Ling, “Three-Dimensional Scattering Center Extraction using the Shooting and Bouncing Ray Technique,” IEEE Transactions on Antennas and Propagation, vol. 44, no. 11, pp. 1445-1453, 1996.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Jianxiong Zhou, Zhiguang Shi, and Qiang Fu, “A Novel Scheme for Global Scattering Center Modeling of Radar Targets,” Frontiers of Electrical and Electronic Engineering in China, vol. 3, pp. 176-182, 2008.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Baiyuan Ding, Ao Zhang, and Rui Li, “A 3-D Scattering Centre Model-Based SAR Target Recognition Method Using Multi-Level Region Matching,” Remote Sensing Letters, vol. 15, no. 3, pp. 215-223, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Ying Xu et al., “SAR Automatic Target Recognition Based on Deep Convolutional Neural Network,” Image and Graphics, 9th International Conference, ICIG 2017, Shanghai, China, vol. 10668, pp. 656-667, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Yang Li et al., “DeepSAR-Net: Deep Convolutional Neural Networks for SAR Target Recognition,” 2017 IEEE 2nd International Conference on Big Data Analysis, Beijing, China, pp. 740-743, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Pengfei Zhao et al., “Multi-Stream Convolutional Neural Network for SAR Automatic Target Recognition,” Remote Sensing, vol. 10, no. 9, pp. 1-22, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Ping Lang et al., “LW-CMDANet: A Novel Attention Network for SAR Automatic Target Recognition,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 15, pp. 6615-6630, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Yikui Zhai et al., “MF-SarNet: Effective CNN with Data Augmentation for SAR Automatic Target Recognition,” The Journal of Engineering, vol. 2019, no. 19, pp. 5813-5818, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Baodai Shi et al., “Synthetic Aperture Radar SAR Image Target Recognition Algorithm based on Attention Mechanism,” IEEE Access, vol. 9, pp. 140512-140524, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Ming Zhang et al., “Convolutional Neural Network with Attention Mechanism for SAR Automatic Target Recognition,” IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Bo Su et al., “CFCANet: A Complete Frequency Channel Attention Network for SAR Image Scene Classification,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 11750-11763, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Chenwei Wang et al., “Multiview Attention CNN-LSTM Network for SAR Automatic Target Recognition,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 12504-12513, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Haohao Ren et al., “Extended Convolutional Capsule Network with Application on SAR Automatic Target Recognition,” Signal Processing, vol. 183, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Jiaqi Shao et al., “A Lightweight Convolutional Neural Network Based on Visual Attention for SAR Image Target Classification,” Sensors, vol. 18, no. 9, pp. 1-21, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Tianwen Zhang, and Xiaoling Zhang, “A Polarization Fusion Network with Geometric Feature Embedding for SAR Ship Classification,” Pattern Recognition, vol. 123, pp. 1-16, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Ronghui Zhan, and Zongyong Cui, “Ship Recognition for SAR Scene Images under Imbalance Data,” Remote Sensing, vol. 14, no. 24, pp. 1-15, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Chenwei Wang et al., “SAR Ship Target Recognition via Multiscale Feature Attention and Adaptive-Weighed Classifier,” IEEE Geoscience and Remote Sensing Letters, vol. 20, pp. 1-5, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Lei Yu et al., “Recognition Method of Ship Target in Complex SAR Image based on Improved ResNet Network,” Systems Engineering and Electronics, vol. 44, no. 12, pp. 3652-3660, 2022.
[CrossRef] [Google Scholar] [Publisher Link]