Unsupervised Polarimetric SAR Image Segmentation and Classification Using Region Growing With Edge Penalty

• Published in: IEEE transactions on geoscience and remote sensing Vol. 50; no. 4; pp. 1302 - 1317
• Main Authors: Yu, P., Qin, A. K., Clausi, D. A.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01-04-2012; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied geophysics; Classification; Complex; Context; Context modeling; Covariance matrix; Earth sciences; Earth, ocean, space; Exact sciences and technology; Image edge detection; Image segmentation; Imagery; Internal geophysics; Iterative methods; Markov processes; Markov random field (MRF); Merging; polarimetry; region adjacency graph (RAG); region-based; Segmentation; Strength; Studies; Synthetic aperture radar; synthetic aperture radar (SAR); Wishart; algorithms; models; polarimetry; extension; strength; accuracy; Markov random field (MRF); classification; Wishart; Complex; amplitude; solution; synthetic aperture radar (SAR); image segmentation; imagery; segmentation; region adjacency graph (RAG); region-based; performances; Synthetic aperture radar
• ISSN: 0196-2892; 1558-0644
• DOI: 10.1109/TGRS.2011.2164085

A region-based unsupervised segmentation and classification algorithm for polarimetric synthetic aperture radar (SAR) imagery that incorporates region growing and a Markov random field edge strength model is designed and implemented. This algorithm is an extension of the successful Iterative Region Growing with Semantics (IRGS) segmentation and classification algorithm, which was designed for amplitude only SAR imagery, to polarimetric data. Polarimetric IRGS (PolarIRGS) extends IRGS by incorporating a polarimetric feature model based on the Wishart distribution and modifying key steps such as initialization, edge strength computation, and the region growing criterion. Like IRGS, PolarIRGS oversegments an image into regions and employs iterative region growing to reduce the size of the solution search space. The incorporation of an edge penalty in the spatial context model improves segmentation performance by preserving segment boundaries that traditional spatial models will smooth over. Evaluation of PolarIRGS with Flevoland fully polarimetric data shows that it improves upon two other recently published techniques in terms of classification accuracy.