Context-based object recognition

Authors:
Shaun Edwards;Meredith Wright;Ben A. Abbott
Affiliations:
Southwest Research Institute, San Antonio, TX;Southwest Research Institute, San Antonio, TX;Southwest Research Institute, San Antonio, TX
Venue:
PerMIS '09 Proceedings of the 9th Workshop on Performance Metrics for Intelligent Systems
Year:
2009

Citing 5
Cited 0

Instance-Based Learning Algorithms

Machine Learning
CYC: a large-scale investment in knowledge infrastructure

Communications of the ACM
Computer Vision

Computer Vision
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
Estimating continuous distributions in Bayesian classifiers

UAI'95 Proceedings of the Eleventh conference on Uncertainty in artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Several methods have been developed for context-based object recognition within aerial imagery. These methods were inspired by human object recognition, which has been shown to rely on contextual information as opposed to classical appearance based methods. While this concept may not be new, this research sought to develop generic methods that leveraged recent developments in cognitive systems research, and more specifically large scale ontologies or knowledge bases. The results of the research have shown that context-based methods, supported by an ontology, can increase recognition rates versus classical appearance based methods. These methods have the potential to automate many complex object recognition tasks, aerial imagery analysis being one of them, that currently require human analysis.