Local Edge Preserving Filter Based Image Fusion and Change Detection in SAR Images Using Fuzzy Clustering Algorithm
Akshara Sasidharan1, Biju V. G2

1Akshara Sasidharan, M. Tech Scholar, Department of Computer Science, College of Engineering Munnar, Idukki, Kerala, India.
2Biju V. G, Assoc. Prof., Department of Electronics and Communication, College of Engineering Munnar, Idukki, Kerala, India.
Manuscript received on June 09, 2015. | Revised Manuscript received on June 11, 2015. | Manuscript published on June 25, 2015. | PP:22-29 | Volume-3 Issue-8, June 2015. | Retrieval Number: H1006063815

Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Change detection plays a vital role in overseeing the transformations on the earth surface. Due to the existence of speckle noise in SAR (Synthetic Aperture Radar) images, change detection becomes a challenging problem. Modified change detection in SAR images is proposed in this paper. It consists of four steps: Difference image generation using Gauss-log ratio operator and Log-ratio operator, image fusion using LEP (Local Edge Preserving) filter, image de-noising using Frost filter and to differentiate changed and unchanged regions in the fused image by using Reformulated Fuzzy Local Information C-Means (RFLICM)clustering. The proposed method is compared with Discrete Wavelet Transform (DWT) based change detection method using synthetic and real SAR images. The performances are evaluated by means of Overall Error (OE), Percentage Correct Classification (PCC), Kappa Coefficient (KC), Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR) and Segmentation Matching Factor (SMF). The results show that combination of gauss-log ratio operator and image fusion using LEP Filtering performs better than dwt based change detection.
Keywords: Change detection, gauss-log ratio, frost filter, local edge preserving (LEP), synthetic aperture radar (SAR).