FUSION OF DEEP LEARNING ARCHITECTURES FOR ENHANCED TARGET RECOGNITION ON SAR IMAGES

(Received: 26-Aug.-2023, Revised: 24-Oct.-2023 , Accepted: 2-Nov.-2023)

Authors K. Cheikh, R. Aitahcene, A. Toumi, Z. Hammoudi,

Keywords #Automatic target recognition #Synthetic aperture radar images #Deep learning #Decision fusion

Abstract In various applications of radar imagery, one of the fundamental problems is mainly linked to the analysis and interpretation of the images provided, in particular the recognition of moving and/or fixed targets. This task has become more difficult due to the large volume of radar data. This led to the use of automatic-processing and target-recognition methods. The aim of this study is to explore data fusion in SAR (Synthetic Aperture Radar) image classifiers. To this end, we propose a new approach to combine three CNN (Convolutional Neural Network) architectures with several fusion rules. First, we perform a training process of three deep-learning architectures; namely, the basic CNN, the Xception and the AlexNet architectures. Then, two fusion techniques are proposed. The first one deals with the majority rule and the second uses a neural network to combine the decision outputs obtained from three elementary classifiers to achieve the final decision. To evaluate and validate the proposed approach, the MSTAR (Moving and Stationary Target Acquisition and Recognition) dataset is used. The obtained performances of the fusion techniques improve the recognition rate with a final accuracy of 99.59% for the majority rule and 99.51% for the neural network-based rule, which surpasses the accuracy of each individual CNN.

References

[1] L. M. Novak, "State-of-the-art of SAR Automatic Target Recognition," Proc. of Record of the IEEE 2000 Int. Radar Conf., pp. 836-843, Alexandria, VA, USA, 2000.

[2] A. Karine, A. Toumi, A. Khenchaf and M. E. Hassouni, "Saliency Attention and Aift Key-points Combination for Automatic Target Recognition on MSTAR Dataset," Proc. of the Int. Conf. on Advanced Technologies for Signal and Image Processing (ATSIP), pp. 1-5, Fez, Morocco, 2017.

[3] D. R. Hush and B. G. Horne, "Progress in Supervised Neural Networks," IEEE Signal Processing Magazine, vol. 10, no. 1, pp. 8-39, 1993.

[4] T. D. Ross et al., "Standard SAR ATR Evaluation Experiments Using the MSTAR Public Release Data Set," Proc. SPIE, Algorithms for Synthetic Aperture Radar Imagery V, vol. 3370, pp.566–573, 1998.

[5] A. Toumi and A. Khenchaf, "Aircrafts Recognition Using Convolutional Neurons Network," Proc. of the Int. Conf. on Radar Systems (RADAR’17), pp.23-26, UK, 2017.

[6] A. Masoomi, R. Hamzehyan and N.C. Shirazi, "Speckle Reduction Approach for SAR Image in Satellite Communication," Int. Journal of Machine Learning and Computing, vol. 2, no. 1, 2012

[7] A. Toumi, A. Alhousseini and A. Khenchef, "Aircraft Recognition Using Convolutional Neurons Network," Proc. of the IEEE 7th Seminar on Detection Systems: Architectures and Technologies (DAT 2017), Algeria, 20-22 February 2017.

[8] Q. Zhao and J. C. Principe, "Support Vector Machines for SAR Automatic Target Recognition," IEEE Transactions on Aerospace and Electronic Systems, vol. 37, pp. 643–654, 2001.

[9] Y. Sun, Z. Liu, S. Todorovic and J. Li, "Adaptive Boosting for SAR Automatic Target Recognition," IEEE Transactions on Aerospace and Electronic Systems, vol. 43, no.1, pp. 112–125, 2007.

[10] A. Karine, A. Toumi, A. Khenchaf and M. Hassouni, "Multivariate Copula Statistical Model and Weighted Sparse Classification for Radar Image Target Recognition," Computers & Electrical Engineering, vol. 84, p. 106633, 2020.

[11] J. C. Cexus, A. Toumi and M. Riahi, "Target Recognition from ISAR Image Using Polar Mapping and Shape Matrix," Proc. of the 2020 IEEE 5th Int. Conf. on Advanced Technologies for Signal and Image Processing (ATSIP), pp. 1-6, Sousse, Tunisia, September 2020.

[12] C. Coman and R. Thaens, "A Deep Learning SAR Target Classification Experiment on MSTAR Dataset," Proc. of the 19th IEEE Int. Radar Symposium (IRS 2018), Bonn, Germany, 20-22 June 2018.

[13] S. A. Wagner, "SAR ATR by a Combination of Convolutional Neural Network and Support Vector Machines," IEEE Trans. on Aerospace and Electronic Systems, vol. 52, no. 6, pp. 2861–2872, 2016.

[14] A. David and E. Morgan "Deep Convolutional Neural Networks for ATR from SAR Imagery," Proc. of SPIE, vol. 9475, DOI: 10.1117/12.2176558, 2015.

[15] A. Ameta, V. Singh1 and V. Devi, "A Convolutional Neural Network Based Approach for SAR Image Classification of Vehicles," Int. Journal of Engineering Research & Technology (IJERT), vol. 9, no. 06, pp. 544-547, 2020.

[16] A. T. Al-Taani and I. T. Al-Dagamesh, " Automatic Detection of Pneumonia Using Concatenated Convolutional Neural Network", Jordanian Journal of Computers and Information Technology (JJICIT), vol. 09, no. 02, pp. 118-136, 2023.

[17] R. Hang, Q. Liu, D. Hong and P. Ghamisi, "Cascaded Recurrent Neural Networks for Hyper-spectral Image Classification," IEEE Trans. Geoscience and Remote Sensing, vol. 57, pp. 5384–5394, 2019.

[18] C. Wang, X. Liu, J. Pei, Y. Huang, Y. Zhang and J. Yang, "Multi-view 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.

[19] X. Fan, L. Chen, X. Xu, C. Yan, J. Fan and X. Li, "Land Cover Classification of Remote Sensing Images Based on Hierarchical Convolutional Recurrent Neural Network," Forests, vol. 14, no. 9, p. 1881, 2023.

[20] Z. Huang, Z. Pan and B. Lei, "Transfer Learning with Deep Convolutional Neural Network for SAR Target Classification with Limited Labeled Data," Remote Sensing, vol. 9, no. 9, p. 907, 2017.

[21] Y. Zhai, H. Ma et al., "SAR ATR with Full-angle Data Augmentation and Feature Polymerization," IET, vol. 19, pp. 6226-6230, 2019.

[22] Z. Geng, Y. Xu, B.-N. Wang, X. Yu, D.-Y. Zhu and G. Zhang, "Target Recognition in SAR Images by Deep Learning with Training Data Augmentation" Sensors, vol. 23, no. 2, p. 941, 2023.

[23] J. Ding, B. Chen, H. Liu and M. Huang, "Convolutional Neural Network with Data Augmentation for SAR Target Recognition," IEEE Geoscience and Remote Sensing Letters, vol. 13, no.3, pp. 364–368, 2016.

[24] L. Lin, J. Li, Q. Yuan and H. Shen, "Polarimetric SAR Image Super-resolution via Deep Convolutional Neural Network", Proc. of the IEEE Int. Geoscience and Remote Sensing Symposium (IGARSS 2019), DOI: 10.1109/IGARSS.2019.8898160, Yokohama, Japan, 2019.

[25] Z. Cui, M. Zhang, Z. Cao and C. Cao, "Image Data Augmentation for SAR Sensor via Generative Adversarial Nets," IEEE Access, vol. 7, pp. 42255 – 42268, 2019.

[26] G. Joshi, R. Natsuaki and A. Hirose "Multi-sensor Satellite-imaging Data Fusion for Earthquake Damage Assessment and the Significant Features," Proc. of the 7th Asia-Pacific Conf. on Synthetic Aperture Radar (APSAR), DOI: 10.1109/APSAR52370.2021.9688438, 2021.

[27] Y. Khenchaf and A. Toumi, "Siamese Neural Network for Automatic Target Recognition Using Synthetic Aperture Radar," Proc. of the Int. Geoscience and Remote Sensing Symposium (IGARSS), Pasadena, USA, Jully 2023.

[28] C. Zhang, Y. Wang, H. Liu, Y. Sun and L. Hu, " SAR Target Recognition Using Only Simulated Data for Training by Hierarchically Combining CNN and Image Similarity," IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022.

[29] Master Public Targets, The Sensor Data Management System, [Online], Available: https://www.sdms.afrl.af.mil/index.php?collection=mstar&page=targets. [30] A. Krizhevsky, I. Sutskever and G. Hinton, "ImageNet Classification with Deep Convolutional Neural Networks," Advances in Neural Information Processing Systems (NIPS), pp. 1-9, 2012.

[31] F. Chollet, "Xception: Deep Learning with Depthwise Separable Convolutions," Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 1251-1258, 2017.

[32] G. Niu, S. S. Lee, B. S. Yang and S. J. Lee, "Decision Fusion System for Fault Diagnosis of Elevator Traction Machine," Journal of Mechanical Science and Technology, vol. 22, no.1, pp. 85-95, 2008.

[33] L. Yu et al., "SAR ATR Based on FCNN and ICAE," Journal of Radars, vol. 7, no.5, pp. 622-631, 2018.

[34] M. Huang, F. Liu and X. Meng, "Few Samples of SAR Automatic Target Recognition Based on Enhanced-shape CNN," Journal of Mathematics, vol. 2021, pp.1-16, 2021.

[35] L. Guo, "SAR Image Classification Based on Multi-feature Fusion Decision Convolutional Neural Network," IET Image Process, vol. 16, pp.1–10, 2022.

[36] J. Pei, Z. Wang et al., "FEF-Net: A Deep Learning Approach to Multi-view SAR Image Target Recognition, " Remote Sensing, vol. 13, no.17, p. 3493, 2021.

[37] S. A. Wagner, "SAR ATR by a Combination of Convolutional Neural Network and Support Vector Machines," IEEE Transactions on Aerospace and Electronic Systems, vol. 52, no. 6, pp. 2861-2872, 2016.