Multiresolution Methods in Systems, Signals, and Images
Abstract
In this final report we summarize our research accomplishments during the support period of this grant. The scope of this research program is to carry out fundamental research in several areas: (a) the use of multiresolution methods in statistically optimal image analysis; (b) the blending of sensor physics and statistical models for computationally efficient, near-optimal inversion and image formation with applications in radar imaging; (c) the development of statistically robust and computationally efficient nonlinear image processing methods with applications in segmentation, edge detection, and feature extraction for object recognition; (d) the development of multiresolution and wavelet-based methods for robust feature extraction, with applications in object recognition; (e) the development of fast numerical methods for a number of difficult problems in statistical image processing; and (f) the extension of our multiresolution modeling framework to include very different granularities of information. from image pixels to discrete variables representing context, with applications to the emerging areas of global awareness and integrated spatial databases.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 30, 2000
- Accession Number
- ADA386019
Entities
People
- Alan S. Willsky
Organizations
- Massachusetts Institute of Technology