Specifications

What is AI?
Super ICEPak
Capabilities
Super ICEPak Uses
Deployment
Specifications
Workshop
Preview Super ICEPak
FAQ
Contact
Product Downloads
Shop Online
What is AI?

Specifications

Classifiers Five Predefined Classifiers: Linear Discriminant Function

bullet

K-Nearest Neighbor

bullet

Empirical Bayesian

bullet

Minimum Distance

bullet

Neural Network

Number of Pattern Classes Supported: unlimited

bullet

Hierarchical multi-level classifier definition setup

bullet

Training/testing set swapping

bullet

Export trained classifier

bullet

Anomaly detection

Features Number of Predefined Waveform Feature Sets: 6

Waveform Signal Feature Domains: 6

bullet

Time Domain

bullet

Power Domain

bullet

Phase Domain

bullet

Cepstral Domain

bullet

Auto-correlation Domain

bullet

Cross-correlation Domain

User-defined feature sets: for image and other applications

Size of feature vector: unlimited

Feature Analysis Aids:

bullet

Automatic Feature Ranking

bullet

Fisher and Bhattacharyya distances

bullet

Feature Probability Density Function Analysis

bullet

3D Feature/Feature Probability Plots

bullet

Density Function Projection

bullet

Feature File/Record Browsing

Operating Modes:

bullet

Visual classifier setup including hierarchical setup

bullet

Classifier Training

bullet

Classifier Evaluation

bullet

Identification of Unknown Signals

Integration and Deployment

The classifier is ready to use with your final target system code mingling: ICEPak's classification and feature extraction engines are self-contained encapsulated modules

bullet

Export feature extraction engine

bullet

Export classification engine

bullet

Rapid prototype creation

bullet

Ready to run sample programs with source code

bullet

Implementation Libraries ANSI C API

bullet

Visual Basic API

bullet

C++ Class library

Data Formats Supported Single byte integer

bullet

Short integer (double byte)

bullet

Long integer (4 bytes)

bullet

Signed and unsigned integer support

bullet

Floating point (4 bytes)

bullet

Double floating point (8 bytes)

bullet

Big-endian and little-endian support

Documentation Extensive on-line HTML reference

bullet

Getting started --- to bring you up to speed in no time

bullet

Tutorial --- guide you through steps of designing your first classifier

bullet

Theory of operation --- everything you want to know about the inner working of the system

bullet

Advanced topics --- information you need to write your own application

 

 

.

 

Back Next

 

Contact:   webmaster@supericepak.com
Copyright © 2002 (TISEC Inc.). All rights reserved.
Revised: November 16, 200
4