Latest News
TISEC delivers
Practical Pattern Recognition Workshop for NDT
in Montreal, Canada, June 28-30, 2010
New look of TISEC's Super ICEPak® web site,
November 2009.
Download and try out Super ICEPak® today, contact us for a trial key,
November 2009.
Super ICEPak® — Specifications
Number of Predefined Classifiers: 5
• Linear Discriminant • K-Nearest Neighbor • Neural Network • Empirical Bayesian • Minimum Distance
Classifier Settings
Number of Pattern Classes Supported: unlimited |
Anomaly detection option |
Hierarchical multi-level classifier definition setup |
Training/testing set swapping |
Export trained classifier and feature extraction engine |
|
Feature Set
Number of Predefined Waveform Feature Sets: 6 |
Size of feature vector: unlimited |
User-defined feature set: for image and other applications |
|
Waveform Signal Feature derived from the following domains
Time |
Power |
Phase |
Cepstrum |
Auto-correlation |
Cross-correlation |
Feature Analysis Aids and Tools
Probability Density Function Projection |
3D Feature/Feature Probability Plots |
Feature Probability Density Function Analysis |
Feature File/Record Browsing |
Fisher and Bhattacharyya Distance Analysis |
Clustering Analysis |
Non-Linear Mapping of Feature Distribution |
Automatic Feature Ranking |
Operating Modes
Visual classifier setup including hierarchical setup |
Classifier Training |
Classifier Evaluation |
Identification of Unknown Signals |
Data Formats Supported
Single byte integer |
Long integer(4 bytes) |
Signed and unsigned integer support |
Short integer(double byte) |
Double floating point(8 bytes) |
Big-endian and little-endian support |
Floating point (4 bytes) |
|
Documentation and Reference
• Getting started — to bring you up to speed in no time
• Theory of operation — everything you want to know about the inner working of the system
• Tutorial — guide you through steps of designing your first classifier
• Advanced topics — information you need to write your own application