Saliency Map Generation for SAR Images with Bayes Theory and Heterogeneous Clutter Model
"> Figure 1
<p>Target and background regions (salient and nonsalient regions) used for the estimation of two probability density functions (PDFs). (<b>a</b>) Regions for the calculation of the local saliency measure; (<b>b</b>) Regions for the calculation of the global saliency measure.</p> "> Figure 2
<p>Framework of the proposed saliency map generation procedure. (<b>a</b>) Synthetic aperture radar (SAR) image saliency map generation; (<b>b</b>) Polarimetric SAR (PolSAR) image saliency generation.</p> "> Figure 3
<p>First study area and single-polarization SAR dataset with X-band. (<b>a</b>) Optical image from Google Earth. The red rectangles denote salient regions through human visual perception; (<b>b</b>) HH polarization SAR image with X-band.</p> "> Figure 4
<p>Saliency map generation results of different methods. (<b>a</b>) The proposed method; (<b>b</b>) The IVSaliency method; (<b>c</b>) The PRSaliency method; (<b>d</b>) Original SAR image overlaid with the salient map of (<b>a</b>), where the red regions are salient.</p> "> Figure 5
<p>The intermediate results of our proposed saliency map generation algorithm. (<b>a</b>) Local saliency map with single scale; (<b>b</b>) Global saliency map with single scale; (<b>c</b>) Combined saliency map with single scale; (<b>d</b>) Multiscale saliency map before refinement; (<b>e</b>) Focus of attention points at single scale; (<b>f</b>) Distance image <math display="inline"> <semantics> <mrow> <msub> <mi>d</mi> <mrow> <mi>foci</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi mathvariant="normal">x</mi> <mo>)</mo> </mrow> </mrow> </semantics> </math> at single scale.</p> "> Figure 6
<p>The study area and Sentinel-1 SAR dataset with C band. (<b>a</b>) Optical image from Google Earth; (<b>b</b>) HH polarization SAR image with C band.</p> "> Figure 7
<p>Saliency map generation results of different methods using the Sentinel-1 dataset. (<b>a</b>) The proposed method; (<b>b</b>) The IVSaliency method; (<b>c</b>) The PRSaliency method; (<b>d</b>) Original SAR image overlaid with the saliency map of (<b>a</b>), where the red regions are salient.</p> "> Figure 8
<p>The study area and Radarsat-2 dataset. (<b>a</b>) Pauli-coded image with C-band (red: HH − VV, green: HV, blue: HH + VV); (<b>b</b>) Optical image from Google Earth, where the red area in the upper left image is the enlarged result of the lower right area.</p> "> Figure 9
<p>Saliency map generation results of Radarsat-2 C-band data with uniform background. (<b>a</b>) The proposed method; (<b>b</b>) The SDPolSAR method; (<b>c</b>) The PRSaliency method; (<b>d</b>) Span image overlaid with the saliency map of (<b>a</b>), where the red regions are salient.</p> "> Figure 10
<p>Third study area and UAVSAR dataset. (<b>a</b>) Pauli-coded image with L-band (red: HH − VV, green: HV, blue: HH + VV); (<b>b</b>) Optical image from Google Earth. The areas with red rectangles represent targets of interest.</p> "> Figure 11
<p>Saliency map generation results of the UAVSAR L-band data with complex image scenes. (<b>a</b>) The proposed method; (<b>b</b>) The SDPolSAR method; (<b>c</b>) The PRSaliency method; (<b>d</b>) Span image overlaid with the saliency map of (<b>a</b>), where the red regions are salient.</p> "> Figure 12
<p>Human-annotated ground truth of salient target detection. (<b>a</b>) X-band SAR image; (<b>b</b>) Sentinel-1 C-band SAR image; (<b>c</b>) Radarsat-2 C-band PolSAR image; (<b>d</b>) UAVSAR L-band PolSAR image.</p> "> Figure 13
<p>Receiver operating characteristic (ROC) curves of various methods using different datasets. (<b>a</b>) X-band SAR image; (<b>b</b>) Sentinel-1 C-band SAR image; (<b>c</b>) Radarsat-2 C-band PolSAR image; (<b>d</b>) UAVSAR L-band PolSAR image.</p> ">
Abstract
:1. Introduction
2. Saliency Indicator for SAR Images
2.1. Local and Global Single-Scale Saliency Measure
2.2. Parameter Estimation of Two PDFs
2.3. Multiscale Saliency Enhancement
2.4. Saliency Refinement Including the Immediate Context
3. Saliency Indicator for PolSAR Images
3.1. PDFs of Salient and Nonsalient Regions in PolSAR Data
3.2. Saliency Indicator for PolSAR Data
4. Experimental Results and Analysis
4.1. Experimental Results with Single-Polarization SAR Data
4.2. Parameter Discussion
4.3. Experimental Results with PolSAR Data
4.4. Performance Evaluation and Comparison
5. Conclusions
Acknowledgments
Author Contributions
Conflicts of Interest
References
- Soergel, U. Radar Remote Sensing of Urban Areas; Springer: Berlin, Germany, 2010. [Google Scholar]
- Lee, J.-S.; Pottier, E. Polarimetric Radar Imaging: From Basics to Applications; CRC Press: Boca Raton, FL, USA, 2009. [Google Scholar]
- Xiang, D.; Tang, T.; Ban, Y.; Su, Y. Man-made target detection from polarimetric SAR data via nonstationarity and asymmetry. IEEE J. Sel. Top. Appl. Earth Obs. 2016, 9, 1459–1469. [Google Scholar] [CrossRef]
- Yu, W.; Wang, Y.; Liu, H.; He, J. Superpixel-based CFAR target detection for high-resolution SAR images. IEEE Geosci. Remote Sens. Lett. 2016, 13, 730–734. [Google Scholar] [CrossRef]
- Tu, S.; Su, Y. Fast and accurate target detection based on multiscale saliency and active contour model for high-resolution SAR images. IEEE Trans. Geosci. Remote Sens. 2016, 54, 5729–5744. [Google Scholar] [CrossRef]
- Zhang, L.; Zou, B.; Tang, W. Similarity-enhanced target detection algorithm using polarimetric sar images. Int. J. Remote Sens. 2012, 33, 6149–6162. [Google Scholar] [CrossRef]
- An, W.; Xie, C.; Yuan, X. An improved iterative censoring scheme for CFAR ship detection with SAR imagery. IEEE Trans. Geosci. Remote Sens. 2014, 52, 4585–4595. [Google Scholar]
- Wu, W.; Guo, H.; Li, X. Urban area man-made target detection for PolSAR data based on a nonzero-mean statistical model. IEEE Geosci. Remote Sens. Lett. 2014, 11, 1782–1786. [Google Scholar] [CrossRef]
- Gao, G. A parzen-window-kernel-based CFAR algorithm for ship detection in SAR images. IEEE Geosci. Remote Sens. Lett. 2011, 8, 557–561. [Google Scholar] [CrossRef]
- Chen, S.; Sato, M. Tsunami damage investigation of built-up areas using multitemporal spaceborne full polarimetric SAR images. IEEE Trans. Geosci. Remote Sens. 2013, 51, 1985–1997. [Google Scholar] [CrossRef]
- Di Bisceglie, M.; Galdi, C. CFAR detection of extended objects in high-resolution SAR images. IEEE Trans. Geosci. Remote Sens. 2005, 43, 833–843. [Google Scholar] [CrossRef]
- Gandhi, P.P.; Kassam, S.A. Analysis of CFAR processors in homogeneous background. IEEE Trans. Aerosp. Electron. Syst. 1988, 24, 427–445. [Google Scholar] [CrossRef]
- Smith, M.E.; Varshney, P.K. VI-CFAR: A novel CFAR algorithm based on data variability. In Proceedings of the IEEE National Radar Conference, Syracuse, NY, USA, 13–15 May 1997; IEEE: Piscataway, NJ, USA, 1997; pp. 263–268. [Google Scholar]
- Kaplan, L.M. Improved SAR target detection via extended fractal features. IEEE Trans. Aerosp. Electron. Syst. 2001, 37, 436–451. [Google Scholar] [CrossRef]
- Tello, M.; López-Martínez, C.; Mallorqui, J.J. A novel algorithm for ship detection in SAR imagery based on the wavelet transform. IEEE Geosci. Remote Sens. Lett. 2005, 2, 201–205. [Google Scholar] [CrossRef]
- Brekke, C.; Anfinsen, S.N.; Larsen, Y. Subband extraction strategies in ship detection with the subaperture crosscorrelation magnitude. IEEE Geosci. Remote Sens. Lett. 2013, 10, 786–790. [Google Scholar] [CrossRef]
- Marino, M.S.-F.; Hajnsek, I.; Ouchi, K. Ship detection with analysis of synthetic aperture radar: A comparison of new and well-known algorithms. Remote Sens. 2015, 7, 5416–5439. [Google Scholar] [CrossRef]
- Iervolino, P.; Guida, R. A novel ship detector based on the generalized-likelihood ratio test for SAR imagery. IEEE J. Sel. Top. Appl. Earth Obs. 2017, 10, 3616–3630. [Google Scholar] [CrossRef]
- Yu, Y.; Wang, B.; Zhang, L. Hebbian-based neural networks for bottom-up visual attention and its applications to ship detection in sar images. Neurocomputing 2011, 74, 2008–2017. [Google Scholar] [CrossRef]
- Tang, T.; Xiang, D.; Xie, H. Multiscale salient region detection and salient map generation for synthetic aperture radar image. J. Appl. Remote Sens. 2014, 8, 083501. [Google Scholar] [CrossRef]
- Wang, H.; Xu, F.; Chen, S. Saliency detector for SAR images based on pattern recurrence. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2016, 9, 2891–2900. [Google Scholar] [CrossRef]
- Zhang, Q.; Wu, Y.; Zhao, W.; Wang, F.; Fan, J.; Li, M. Multiple-scale salient-region detection of SAR image based on gamma distribution and local intensity variation. IEEE Geosci. Remote Sens. Lett. 2014, 11, 1370–1374. [Google Scholar] [CrossRef]
- Zhang, Q.; Wu, Y.; Wang, F.; Fan, J.; Zhang, L.; Jiao, L. Anisotropic-scale-space-based salient-region detection for SAR images. IEEE Geosci. Remote Sens. Lett. 2016, 13, 457–461. [Google Scholar] [CrossRef]
- Zhang, X.; Meng, H.; Ma, Z.; Tian, X. SAR image despeckling by combining saliency map and threshold selection. Int. J. Remote Sens. 2013, 34, 7854–7873. [Google Scholar] [CrossRef]
- Zheng, Y.; Jiao, L.; Liu, H.; Zhang, X.; Hou, B.; Wang, S. Unsupervised saliency-guided SAR image change detection. Pattern Recognit. 2017, 61, 309–326. [Google Scholar] [CrossRef]
- Itti, L.; Koch, C.; Niebur, E. A model of saliency-based visual attention for rapid scene analysis. IEEE Trans. Pattern Anal. Mach. Intell. 1998, 20, 1254–1259. [Google Scholar] [CrossRef]
- Harel, J.; Koch, C.; Perona, P. Graph-based visual saliency. In Proceedings of the 20th Annual Conference on Neural Information Processing Systems, Vancouver, BC, Canada, 4–7 December 2006; pp. 545–552. [Google Scholar]
- Jin, S.; Wang, S.; Li, X.; Jiao, L.; Zhang, J.A.; Shen, D. A salient region detection and pattern matching-based algorithm for center detection of a partially covered tropical cyclone in a SAR image. IEEE Trans. Geosci. Remote Sens. 2017, 55, 280–291. [Google Scholar] [CrossRef]
- Jager, M.; Hellwich, O. Saliency and salient region detection in SAR polarimetry. In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, Seoul, Korea, 29–29 July 2005; pp. 2791–2794. [Google Scholar]
- Huang, X.; Huang, P.; Dong, L.; Song, H.; Yang, W. Saliency detection based on distance between patches in polarimetric SAR images. In Proceedings of the International Geoscience and Remote Sensing Symposium, Quebec City, QC, Canada, 13–18 July 2014; pp. 4572–4575. [Google Scholar]
- Gao, G. Statistical modeling of SAR images: A survey. Sensors 2010, 10, 775–795. [Google Scholar] [CrossRef] [PubMed]
- Deng, X.; López-Martínez, C.; Chen, J.; Han, P. Statistical modeling of polarimetric SAR data: A survey and challenges. Remote Sens. 2017, 9, 348. [Google Scholar] [CrossRef]
- Marques, R.C.P.; Medeiros, F.N.; Nobre, J.S. SAR image segmentation based on level set approach and G0 model. IEEE Trans. Pattern Anal. Mach. Intell. 2012, 34, 2046–2057. [Google Scholar] [CrossRef] [PubMed]
- Frery, A.C.; Muller, H.-J.; Yanasse, C.C.F.; Sant’Anna, S.J.S. A model for extremely heterogeneous clutter. IEEE Trans. Geosci. Remote Sens. 1997, 35, 648–659. [Google Scholar] [CrossRef]
- Goferman, S.; Zelnik-Manor, L.; Tal, A. Context-aware saliency detection. IEEE Trans. Pattern Anal. Mach. Intell. 2012, 34, 1915–1926. [Google Scholar] [CrossRef] [PubMed]
- Freitas, C.C.; Frery, A.C.; Correia, A.H. The polarimetric G distribution for SAR data analysis. Environmetrics 2005, 16, 13–31. [Google Scholar] [CrossRef]
X Band SAR Image | Sentinel-1 C Band SAR Image | Radarsat-2 C Band Image | UAVSAR L Band Image | |
---|---|---|---|---|
The proposed method | 0.9756 | 0.9833 | 0.9879 | 0.9512 |
The IVSaliency method | 0.9347 | 0.7241 | ||
The PRSaliency method | 0.8435 | 0.8012 | 0.7651 | 0.9323 |
The SDPolSAR method | 0.9014 | 0.7345 |
X-Band SAR Image 300 × 500 | Sentinel-1 C-Band SAR Image 800 × 700 | Radarsat-2 C-Band Image 300 × 300 | UAVSAR L-Band Image 250 × 600 | |
---|---|---|---|---|
The proposed method | 20.14 | 52.36 | 28.54 | 36.15 |
The IVSaliency method | 13.25 | 30.19 | ||
The PRSaliency method | 12.14 | 26.92 | 10.38 | 18.57 |
The SDPolSAR method | 18.17 | 25.78 |
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Xiang, D.; Tang, T.; Ni, W.; Zhang, H.; Lei, W. Saliency Map Generation for SAR Images with Bayes Theory and Heterogeneous Clutter Model. Remote Sens. 2017, 9, 1290. https://doi.org/10.3390/rs9121290
Xiang D, Tang T, Ni W, Zhang H, Lei W. Saliency Map Generation for SAR Images with Bayes Theory and Heterogeneous Clutter Model. Remote Sensing. 2017; 9(12):1290. https://doi.org/10.3390/rs9121290
Chicago/Turabian StyleXiang, Deliang, Tao Tang, Weiping Ni, Han Zhang, and Wentai Lei. 2017. "Saliency Map Generation for SAR Images with Bayes Theory and Heterogeneous Clutter Model" Remote Sensing 9, no. 12: 1290. https://doi.org/10.3390/rs9121290