Comparison of Various Speckle Noise Reduction Filters on Synthetic Aperture Radar Image

• Main Authors: Santoso, Ardhi W., Bayuaji, Luhur, Lim, Tien Sze, Habibah, Lateh, Jasni, Mohamad Zain
• Format: Article
• Language: English; English
• Published: Research India Publications 2016
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://umpir.ump.edu.my/id/eprint/14342/1/fskkp-2016-santoso-Comparison%20of%20various%20speckle%20noise%20reduction.pdf; http://umpir.ump.edu.my/id/eprint/14342/7/fskkp-2016-santoso-Comparison%20of%20various%20speckle%20noise%20reduction1.pdf; http://umpir.ump.edu.my/id/eprint/14342/; http://www.ripublication.com/ijaer16/ijaerv11n15_61.pdf
• Institution: Universiti Malaysia Pahang

Synthetic Aperture Radar (SAR) image with its advantages, is becoming popular than the optical image in earth observation using the remote - sensing techniques. The SAR image has a high resolution and not influenced by weather conditions either day or night. SAR's image formation process led to speckle noise; it causes difficulties during the process of interpretation and analysis of SAR images. Thus, speckle noise reduction needs to be deployed prior to the use of the SAR images. The ideal speckle filter has the capability of reducing speckle noise without losing the information and content, while preserving the edges and features. To date, various no ise filters have been designed for different purposes and different capacities. In this study, we discussed four filters, namely Lee, Frost, Median and Mean filter. Those four filters are analyzed and compared based upon the quality parameter and statistic al performance using SAR sample image respectively. We are analyzing quality parameter and comparing statistical performance of Lee, Frost, Mean and Median filters for SAR sample image. The results show Mean Square Error (MSE), Average Difference (AD), Pea k Signal to Noise Ratio (PSNR), Signal to Noise Ratio (SNR) and Structural Similarity Index Measure (SSIM) value that generated on SAR image with four different areas by Frost filter performs better than the other filter. Visual interpretation of the de - sp eckle image that filtered with Frost filter shows sharpens edge and preserved texture to the SAR image