Build AI and classifier applications fast!
 
  • Neural Networks
  • Self-Organizing Map
  • Naive Bayes
  • K Nearest Neighbor
  • Back Propagation
  • Data Preparation

 

What can I build with IntelligenceLab?

IntelligenceLab is an Artificial Intelligence library that can be used for data grouping, SPAM filters, computer vision, speech recognition, or any other classification task.
 
The library contains classifiers, training components, and a general purpose neuron component.
 
IntelligenceLab is implemented using very fast MMX optimized libraries.
 
Intel:
  • Intel MMX
  • Intel Performance Primitives ( IPP )
 
 
IntelligenceLab is fully compatible and directly integrates with SignalLab, VideoLab, AudioLab, VisionLab, InstrumentLab, and PlotLab.
Delphi 2010 C++ Builder 2010 RAD Studio 2010 Delphi Prism

Classifiers

ILNaiveBayes ILNaiveBayes Naive Bayesian classifier
ILKNearestNeighbor ILKNearestNeighbor

Nearest Neighbor classifier

ILNeuralNetwork ILNeuralNetwork Neural Network classifier
ILSelfOrganizingMap ILSelfOrganizingMap Self Organizing Map classifier
ILRadialBasisFunctionNetwork ILRadialBasisFunctionNetwork Radial Basis Function Network classifier.

Converters

SLPair SLPair Pairs multiple data buffers together

Custom Filters

SLGenericReal SLGenericReal User defined filter for Real data
SLGenericFilter SLGenericFilter User defined filter for Binary data
SLGenericRealMatrix SLGenericRealMatrix User defined generator for Real matrix data

Generic Filters

SLGenericReal SLGenericReal User defined filter for Real data
SLGenericFilter SLGenericFilter User defined filter for Binary data
SLGenericRealMatrix SLGenericRealMatrix User defined generator for Real matrix data

Generic Generators

SLGenericGen SLGenericGen User defined generator for Binary data
SLGenericRealGen SLGenericRealGen User defined generator for Real data

Other

ILNeuron ILNeuron General purpose neuron component

Signal Filters

SLGenericReal SLGenericReal User defined filter for Real data
SLGenericFilter SLGenericFilter User defined filter for Binary data
SLGenericRealMatrix SLGenericRealMatrix User defined generator for Real matrix data

Signal Generators

SLGenericGen SLGenericGen User defined generator for Binary data
SLGenericRealGen SLGenericRealGen User defined generator for Real data

Training

ILNeuralNetworkBackpropTrain ILNeuralNetworkBackpropTrain Trains neural networks using Backprop algorithm
ILNeuralNetworkRPropTrain ILNeuralNetworkRPropTrain Trains neural networks using RProp algorithm
ILTrainingData ILTrainingData Prepares training data for the classifiers

IntelligenceLab: What's New

New in Version 5.0.1 (released on 2011-10-18)

  • The managed assemblies are now rewritten in C# for better .NET 4.0 compatibility.
  • Added support for .NET 4.0 and .NET 3.5 “Client Profile”.
  • Significantly improved .NET type converters.
  • Significantly optimized component loading performance in .NET.
  • Major improvements in the .NET property editors.
  • Small improvements in the Unicode support for VC++.
  • Improved VisualStudio 2010 .NET support.
  • .NET assemblies have been renamed to better match the Microsoft guidelines.
  • Added .NET 4.0 assemblies.

New in Version 5.0 (released on 2011-09-05)

  • Added Delphi XE and XE2 support.
  • Added TILRadialBasisFunctionNetwork component.
  • Training can be saved and restored.
  • Neural network training progress can be monitored.
  • Significantly improved multithreading model adapted for multicore systems, with advanced relaxed interlocking, and with optional per-component dedicated threads.
  • Added new demos.

New in Version 4.5

This is the first official IntelligenceLab release.

( Click to view Larger images. )
These demos are executable after downloading.
Also see:   Demo source code downloads

IntelligenceLab Examples

Self Organizing Map classifier Self Organizing Map classifier
Naive Bayes classifier Naive Bayes classifier
K Nearest Neighbor classifier K Nearest Neighbor classifier
Neural Network classifier Neural Network classifier
Face Recognition Demo Face Recognition Demo

IntelligenceLab: Frequently Asked Questions

Q: Do I have to purchase IntelligenceLab?
A: InstrumentLab is free for non commercial purposes. This means that you can use it for personal pleasure, hobby, or evaluation purposes. If you are student you can also use it for free for your educational needs including school projects. If you are a tutor at present you also can use it for free for your classes. However in the near future a low cost tutor license will be available, and you will have to purchase it. Your students however will not need license to use the product at class.
Q: Is IntelligenceLab multithreading? Will it stop streaming while I move the window with the mouse?
A: Yes. IntelligenceLab is fully multithreaded and will not stop streaming while the main thread is busy.

IntelligenceLab Installers

VCL Installer: IntelligenceLab 5.0.1 for Delphi and C++ Builder
Updated on 2011-10-18, downloaded 2082 times.
VCL Installer: IntelligenceLab 5.0.1 for Delphi and C++ Builder -- XE2 Update 2 Only
Updated on 2011-11-04, downloaded 81 times.
VCL Installer: IntelligenceLab 5.0.1 FireMonkey (Windows only) Beta 1 for XE2 Update 2 Only
Updated on 2011-11-12, downloaded 81 times.
Visual C++ Installer: IntelligenceLab 5.0.1 for Visual C++ MFC
Updated on 2011-10-18, downloaded 404 times.
.NET Installer: IntelligenceLab 5.0.1 for Microsoft .NET 2.0-4.0
Updated on 2011-10-18, downloaded 871 times.

Demos

These demos are executable after downloading.   Also see: Demo source code downloads
K-Nearest Neighbors demo

Demonstrates a machine learning algorithm for classifying objects

Neural Network demo

Demonstrates a neural network classifier

Self-Organizing Map demo

Demonstrates a type of neural network that uses unsupervised learning

Naive Bayes demo
Demonstrates a Naive Bayes classifier
Face Recognition Demo

Demonstrates training a Radial Basis Function Network to recognize faces, and then performing the actual recognition.

 

The libraries are licensed per developer, and there are no royalties required. Once you have purchased a developer license, you are free to distribute your final product without the need to pay any royalties or other fees as long as you comply with the following license agreement:
 
 
LICENSE - AGREEMENT
 
Users of the LabPacks Component Pack(Software) must accept these terms of license.
 
If you refuse to accept the terms of license, you may not use this software, and you have the right to return it within 30 days of the purchase date, and to receive your money back.
 
For your convenience a copy of this license agreement will be stored on your system during the installation.
 
This license grants you the following rights:
 
You have a non exclusive license to the Software. The Title and all intangible rights to the Software are the property of Mitov Software .  
 
You may install and use one copy of the Software per computer assuming only one and the same person will use the Software on all of the computers. You may distribute any application that includes Software without any additional royalties beyond your initial license registration fee.
 
You can also create a reasonable set of copies of the product on different types of media, such as CD Rom or Type Backup, as far as those copies are used only for your own back up protection.
 
Description of the limitations. 
 
You may not reverse engineer, decompile, or disassemble the Software.
 
The Software is licensed as a single product. You may not rent or lease the Software. You must treat the Software like any other copyrighted material except that you may either (a) a reasonable amount of copies of the Software solely for backup or archival purposes, or (b) install the Software multiple computers provided you keep the original solely for backup or archival purposes and only a single developer will use all of the copies.
 
You may not separate Delphi and C++ Builder components from this Software.
Each developer must have a valid license for the Software.
 
You may not create a competing component product using this Software.
 
Additional restriction to Delphi-VCL and C++ Builder components: 
 
You may not distribute the Delphi-VCL components in a DCU (Delphi Compiled Unit) format. You can distribute the component only as a part of your executable or dynamic link library program in a compiled form.
 
You may not distribute the C++ Builder components in OBJ format. You can distribute the component only as a part of your executable or dynamic link library program in a compiled form.
 
 
Mitov Software grants a limited warranty with these restrictions:
 
THE SOFTWARE IS SUPPLIED AS IT IS. MITOV SOFTWARE AND ITS SUPPLIERS DISCLAIM ALL WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WITH REGARD TO THE SOFTWARE. NO LIABILITY FOR CONSEQUENTIAL DAMAGES. TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT SHALL MITOV SOFTWARE OR ITS SUPPLIERS BE LIABLE FOR ANY SPECIAL, INCIDENTAL, INDIRECT, OR CONSEQUENTIAL DAMAGES WHATSOEVER (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF BUSINESS PROFITS, BUSINESS INTERRUPTION, LOSS OF BUSINESS INFORMATION, OR ANY OTHER PECUNIARY LOSS) ARISING OUT OF THE USE OF OR INABILITY TO USE THE SOFTWARE.

News

November 12, 2011

FireMonkey Beta 1 release!

October 26, 2011

The NEW OpenWire Editor version 5.0.1 beta is officially released

October 18, 2011

Release of version 5.0.1 of the libraries on all platforms

October 12, 2011

And don't miss my second talk: Video, Audio Processing, DSP, Artificial Intelligence and Computer Vision with FireMonkey

October 11, 2011

Don't miss my speech at CodeRage 6: Creating dual-compile controls for VCL and FireMonkey

September 2, 2011

Welcome to our New Web Site!

read more

Partners

Em