Learning Integrated Visual Database for Image Exploitation
Abstract
The research summarized in this report is aimed at developing image understanding (IU) algorithms and systems that have performance prediction and learning capabilities and that can improve their performance with experience, in terms of quality of results, processing speed and matching with the user's perception. The following scientific problems are addressed: (a) Fundamental theory for predicting the performance of object recognition systems and its validation on SAR images, (b) Automatic methods for recognizing articulated, occluded and configuration variants of targets in SAR images and video, (c) Adaptive learning integrated target recognition algorithms/systems, and (d) Learning visual concepts in images/videos with user interaction and experience over time. The research presented makes a significant contribution to real-world applications which require robust high performance automated systems that can recognize objects in reconnaissance imagery acquired under dynamically changing conditions and for systems that can efficiently extract meaningful information from enormous image/video databases.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 25, 2002
- Accession Number
- ADA413389
Entities
People
- Bir Bhanu
Organizations
- University of California, Riverside