Automatic detection of geomorphology changes using multisar data

Abstract

This study presents work to utilize multi Synthetic Aperture Radar (SAR) data to map the rate of spit change along the east coast of Kuala Terengganu, Malaysia. Multi temporal SAR data are involved the spacebome and airborne SAR data. The spacebome are ERS-I, and RADARS AT -1 SAR data. Two algorithms have implemented to extract the spit ‘s boundary from multi-SAR data which are Lee and Enhance Lee algorithms. The two algorithms are applied to multi-SAR data with window kernel sizes of 3x3, 5x5, and 7x7, respectively. The rate of spit change is estimated by using the traditional method of historical vectors overlaying technique which were acquired from multi-SAR data by manual digitizing. This study shows that Enhance Lee algorithm with kernel window size of 5 x 5 provides better speckles reduction. It is interesting too to find that the spit has dominated by accretion since 1993 to 2003 which is 7 m/yr. It can be concluded that the multi- SAR data can be used for mapping coastal geomorphology change in tropical area such as Malaysia.