| Author | Jayasekera, Champika W. |
| Call Number | AIT Thesis no. CS-92-06 |
| Subject(s) | Image processing--Digital techniques--Remote sensing
|
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of
Master of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. CS-92-06 |
| Abstract | The presence of speckle in radar images makes the radiometric and textural aspects less
efficient for class discrimination. Many adaptive filters have been developed for speckle
reduction.
In this study, the most well-known and promising filters have been analyzed. Both
conventional and exclusively SAR filters have been studied. The latter category of filters are
based on a test related to the local coefficient of variation of the observed image, which describes
the scene heterogeneity. Some modifications have then been introduced to the Refined GammaMap filter in order to improve its filtering effectiveness. This has stemmed from a proposed
refined speckle filtering algorithm for SAR images which ensures reliable local statistics
estimation, edge enhancement and preservation of both textural and curvilinear features. In these
modifications the determination of the speckle coefficient of variation and the orientation of a
possible edge or linear feature have been carried out in a more practical and holistic manner to
avert the imperfections in the existing filter.
All the filters have been tested on ERS-1 SAR image taken in Flevoland, Netherlands and
Kanchanaburi,Thailand. As was expected, the new filter performs better, i.e.,it average the
homogeneous areas better and preserve texture information, edges, linear features, and point
target responses finely at the mean time. Moreover, this filter can be adapted to features other
than the coefficient of variation to reduce speckles whilst preserving the corresponding textural
and structural features. |
| Year | 1992 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. CS-92-06 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Hosomura, Tsukasa; |
| Examination Committee(s) | Aschbacher, Josef ;Yulu, Qi ; |
| Scholarship Donor(s) | Norwegian Agency for Development and Corporation (NORAD) ; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 1992 |