Recognition Toolkit
The Recognition Toolkit provides a set of tasks for object recognition and classification. Tasks use probabilistic analysis and information measures to select attributes and features for classifier development.
- Different automatic probabilistic and information based classifiers can be generated that do not require the user to determine complex rules for object recognition
- Symbolic information can be processed as readily a standard measurement data
- Demonstration detection, training, and recognition graphs are provided that can be directly employed or used as templates.
The Recognition Toolkit provides a capability for pattern recognition on continuous measurement data, discrete data or symbolic information. It contains a series of tasks for classification, training, feature evaluation, clustering, coding and extraction. Twenty tasks form the pattern recognition core. Five additional advanced extraction and demonstration tasks provide assistance with complex recognition development.

Example of parameter-free determination of natural classes for scene recognition

An example of land use classifiication from multi-spectral imagery
The core task set includes four classifiers, four training tasks, an automatic natural clustering routine, five feature evaluation and analysis tasks, and six utility tasks for preprocessing and building multistage alternative classifiers.

KBVision Modules
(webmaster@aai.com)
KBVision is a trademark of Amerinex Artificial Intelligence, Inc.