An improved fast level set method initialized with a combination of k-means clustering and Otsu thresholding for unsuper
- PDF / 3,831,309 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 71 Downloads / 178 Views
ORIGINAL PAPER
An improved fast level set method initialized with a combination of k-means clustering and Otsu thresholding for unsupervised change detection from SAR images Armin Moghimi 1 & Safa Khazai 2 & Ali Mohammadzadeh 1
Received: 18 October 2016 / Accepted: 12 June 2017 # Saudi Society for Geosciences 2017
Abstract Detection of changes in synthetic aperture radar (SAR) images is an important challenge due to the effects of speckle noise on these images. In recent years, appropriate methods for SAR-based-change detection have been developed based on the level set methods (LSM). These methods need to set parameters for defining a proper initial contour. Moreover, the gradient information is only employed in the total energy of these methods for segmentation of the difference image. In this study, a novel method has been proposed for unsupervised change detection of multitemporal SAR images based on the improved fast level set method (IFLSM) initialized with a combination of kmeans and Otsu techniques. The proposed method utilizes the discrete wavelet transform (DWT) fusion strategy and edge enhancement to achieve a noise-resistant difference image from the mean-ratio and log-ratio images. Afterward, the generated binary change map (CM) by applying a combination of k-means and Otsu techniques on the difference image is used as the initial contour to achieve a final CM on difference image using the IFLSM. To check advantages of the proposed method, experiments are applied on two sets of multitemporal SAR images corresponding to artificial Chitgar Lake (under reconstruction)
* Ali Mohammadzadeh [email protected] Armin Moghimi [email protected] Safa Khazai [email protected] 1
Department of the Photogrammetry and Remote Sensing, Geomatics and Geodesy Engineering Faculty, K.N.Toosi University of Technology, Tehran, Iran
2
Civil Engineering Research Center, Imam Hussein Comprehensive University, Tehran, Iran
in Tehran (Iran) taken by TerraSAR-X satellite in 2011 and 2012, and corresponding to San Pablo and Briones reservoirs in California (USA) acquired by ERS-2 satellite in 2003 and 2004. Results of proposed method were compared with results of some well-known unsupervised change detection methods. Experimental results prove the sufficiency of the proposed method in unsupervised change detection in terms of accuracy, implementation time, and computational complexity. Keywords Unsupervised change detection . K-means clustering . Otsu thresholding . Level set method . Multitemporal SAR images
Introduction Monitoring environmental changes is one of the main concerns for geoscientists and researchers of related sciences. The optimal exploitation of earth’s natural resources needs to a deep understanding of the logical relationship between human and natural phenomena and also their interactions. This can be provided by identifying differences between phenomena in the same geographical area at different times by observing them which is called change detection of earth’s surface (Lu et al. 2004; Singh 1989).
Data Loading...
