Unsupervised SAR Image Change Detection Based on SIFT Keypoints and Region Information

This letter presents a new unsupervised distribution-free change detection method for synthetic aperture radar (SAR) images based on scale-invariant feature transform (SIFT) keypoints and region information. Since the SIFT can detect blob-like structures in an image and be insensitive to noise, we first extract noise-robust SIFT keypoints in the log-ratio image to reduce the detection range. Then, in order to obtain accurate changed regions, rather than directly obtaining the change-detection map from the difference image as in some traditional change detection methods, we make segmentation around the extracted keypoints in the two original multitemporal SAR images, where the edges of detection regions are much clearer than those in the difference image, and further compare the two segmentations to generate the change-detection map. This method utilizes the blob-like structure information offered by SIFT keypoints and the region information extracted via image segmentation. Experiments on real SAR images demonstrate the effectiveness of the proposed method.

[1]  Peng Zhang,et al.  Semisupervised SAR Image Change Detection Using a Cluster-Neighborhood Kernel , 2014, IEEE Geoscience and Remote Sensing Letters.

[2]  Gabriele Moser,et al.  Generalized minimum-error thresholding for unsupervised change detection from SAR amplitude imagery , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[3]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[4]  Jean-Yves Tourneret,et al.  Change Detection in Multisensor SAR Images Using Bivariate Gamma Distributions , 2008, IEEE Transactions on Image Processing.

[5]  Yifang Ban,et al.  Unsupervised Change Detection in Multitemporal SAR Images Over Large Urban Areas , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[6]  Lorenzo Bruzzone,et al.  An unsupervised approach based on the generalized Gaussian model to automatic change detection in multitemporal SAR images , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[7]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[8]  Krystian Mikolajczyk,et al.  Detection of local features invariant to affines transformations , 2002 .

[9]  Tony Lindeberg,et al.  Detecting salient blob-like image structures and their scales with a scale-space primal sketch: A method for focus-of-attention , 1993, International Journal of Computer Vision.

[10]  Liujuan Cao,et al.  Oil spill detection based on a superpixel segmentation method for SAR image , 2014, 2014 IEEE Geoscience and Remote Sensing Symposium.

[11]  P. Lombardo,et al.  Maximum likelihood approach to the detection of changes between multitemporal SAR images , 2001 .

[12]  D Marr,et al.  Theory of edge detection , 1979, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[13]  Cordelia Schmid,et al.  A performance evaluation of local descriptors , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Turgay Çelik,et al.  Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and $k$-Means Clustering , 2009, IEEE Geoscience and Remote Sensing Letters.

[15]  Ashish Ghosh,et al.  Fuzzy clustering algorithms for unsupervised change detection in remote sensing images , 2011, Inf. Sci..

[16]  Maoguo Gong,et al.  Change Detection in Synthetic Aperture Radar Images based on Image Fusion and Fuzzy Clustering , 2012, IEEE Transactions on Image Processing.

[17]  Yong Wang,et al.  Building identification from SAR image based on the modified marker-controlled watershed algorithm , 2015, 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[18]  Shuang Wang,et al.  Unsupervised Change Detection in SAR Image Based on Gauss-Log Ratio Image Fusion and Compressed Projection , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.