Call For Papers
Contact Us

  SAR Image Target Classification: A Feature Fusion Approach  
  Authors : Sivaranjani Rajamanickam; S.Mohamed Mansoor Roomi
  Cite as:


Identifying and recognizing vehicles in Synthetic Aperture Radar (SAR) images are key for military application. This paper presents a thorough exploratory work on SAR image target classification utilizing feature fusion strategy. The combination of features is examined with respect to their classification accuracy. The test SAR image is processed by a SAR-BM3D filter to remove speckle noise. Then the salient region of the image is extracted using context aware saliency detection model to detect the potential regions of interest (ROI) which reduces the search space. The different texture characteristic values of GLCM are computed and twenty geometrical features such as centroid, area are calculated for ROI. The features are cascaded and applied to a popular classifier such as Support Vector Machines (SVM) and K-Nearest Neighbor (KNN). Experimental results shown on a MSTAR SAR imagery dataset for three classes exhibit the superior performance of the proposed methods.


Published In : IJCSN Journal Volume 6, Issue 6

Date of Publication : December 2017

Pages : 689-693

Figures :05

Tables : 01


Sivaranjani Rajamanickam : Dept of ECE,Sethu Institute of Technology Virudhunagar, India

S.Mohamed Mansoor Roomi : Dept of ECE, Thiagarjar College of Engineering Madurai, India.


GLCM, geometrical feature, saliency, Classification, True positive rate

In this work, a feature fusion based classification framework on MSTAR dataset for classifying military target in SAR image is proposed. The proposed method applies cascaded features such as GLCM texture feature and shape features. These features are trained by a SVM and KNN classifier. From the experimentation it is clearly evident that fused feature based classification provides better results compared to their individual performance. At the same time SVM based classification provides better result than KNN interms of classification accuracy and true positive rate.


[1] Gonzalez R C and Woods R E “ Digital image processing. Upper Saddle River”, NJ: Pearson Education. 2009 [2] F. Argenti, A. Lapini, L. Alparone, and T. Bianchi, “A Tutorial on Speckle Reduction in Synthetic Aperture Radar Images,” IEEE Geosci. Remote Sens. Mag., no. September, pp. 6–35, 2013. [3] K. El-Darymli, E. W. Gill, P. McGuire, D. Power, and C. Moloney, “Automatic Target Recognition in Synthetic Aperture Radar Imagery: A State-of-the-Art Review,” IEEE Access, vol. 4, pp. 6014–6058, 2016. [4] B. Xu, Y. Cui, Z. Li, B. Zuo, J. Yang, and J. Song, “Patch Ordering-Based SAR Image Despeckling Via Transform-Domain Filtering,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 8, no. 4, pp. 1682–1695, 2015. [5] J. Ai, X. Qi, W. Yu, Y. Deng, F. Liu, and L. Shi, “A new CFAR ship detection algorithm based on 2-D joint lognormal distribution in SAR images,” IEEE Geosci. Remote Sens. Lett., vol. 7, no. 4, pp. 806–810, 2010. [6] S. Banerjee, N. Gupta, S. Das, and P. R. O. Y. Chowdhury, “Detecting aircrafts from satellite images using saliency and conical pyramid based template representation,” vol. 41, no. 10, pp. 1155–1171, 2016. [7] R. Ressel, A. Frost, and S. Lehner, “A Neural NetworkBased Classification for Sea Ice Types on X-Band SAR Images,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 8, no. 7, pp. 3672–3680, 2015. [8] H. Liu, H. Guo, and L. Zhang, “SVM-Based Sea Ice Classification Using Textural Features and Concentration From RADARSAT-2 Dual-Pol ScanSAR Data,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 8, no. 4, pp. 1601–1613, 2015. [9] Goferman S, Zelnik-Manor L and Tal A ,” Contextaware saliency detection” IEEE Trans. Pattern Anal. Mach. Intell. 34(10): 1915–1926,2012 [10] G. Dong and G. Kuang, “Classification on the monogenic scale-space: application to target recognition in SAR image,” IEEE Trans. Image Process., vol. 24, no. 8, pp. 1–1, 2015. .[11] T. Cooke, “Detection and Classification of Objects in Synthetic Aperture Radar Imagery,” Sci. Technol., 2006. [12] S. Qi, J. Ma, J. Lin, Y. Li, and J. Tian, “Unsupervised Ship Detection Based on Saliency and S-HOG Descriptor From Optical Satellite Images,” IEEE Geosci. Remote Sens. Lett., vol. 12, no. 7, 2015. [13] G. Liu and H. Zhong, “Nonlocal means filter for polarimetric SAR data despeckling based on discriminative similarity measure,” IEEE Geosci. Remote Sens. Lett., vol. 11, no. 2, pp. 514–518, 2014. [14] B. Bhanu, “Automatic Target Recognition: State of the Art Survey,” IEEE Trans. Aerosp. Electron. Syst., vol. AES-22, no. 4, pp. 364–379, 1986. [15] J. C. Principe, M. Kim, and J. W. Fisher III, “Target discrimination in synthetic aperture radar using artificial neural networks.,” IEEE Trans. Image Process., vol. 7, no. 8, pp. 1136–49, 1998.