Abstract
A multi-scale texture segmentation algorithm for SAR images based on the discrete wavelet transform is presented. Responses from different sub-bands are used to form a feature vector for each pixel position that is the input to the classification scheme. To further improve the classification results, the tree-structured wavelet packet transform is used to automatically identify a suitable sub-set of elements from the feature vector. Results show that this approach is effective for both the redundant and non-redundant wavelet transforms.
Original language | English |
---|---|
Pages | III-1438-III-1440 |
DOIs | |
Publication status | Published - 24 Jun 2002 |
Event | IEEE International Geoscience and Remote Sensing Symposium (IGARSS '02) - Toronto, Canada Duration: 24 Jun 2002 → 28 Jun 2002 |
Conference
Conference | IEEE International Geoscience and Remote Sensing Symposium (IGARSS '02) |
---|---|
Country/Territory | Canada |
City | Toronto |
Period | 24/06/02 → 28/06/02 |
Keywords
- Geophysical signal processing
- Image texture
- Radar imaging
- Feature vector
- Multi-scale texture segmentation algorithm
- Image segmentation
- Remote sensing by radar
- Redundant wavelet transforms
- SAR images
- Discrete wavelet transforms
- Minimum distance texture classification
- Synthetic aperture radar
- Sub-bands tree-structured wavelet packet transform
- Image classification
- Discrete wavelet transform
- Wavelet packet
- Nonredundant wavelet transforms