about ThPP
download
requirements
algorithms and samples
license
SourceForge Logo
thermal processing panel

about
ThPP is a processing and analysis tool for infrared images. These infrared images are obtained with the use of thermography cameras.
Supported cameras:
devicetypesupporteddescription
IRTIS200yesstill 256x248

For now, only IRTIS images are supported, but this will be subject to change. If a manufacturer of thermal cameras is interested in making this software work with it's own devices please contact me.

download
ThPP is available as a Windows i586 binary and as Borland C++ Builder source code.
You can also download a package containing 10 example IR images.
The project's documentation is in a separate file and it contains short descriptions of the available functions and sample images. All of the above can be found on the sourceforge download page.

requirements
OSworksobservations
Windows 95yesno problems
Windows 98yesno problems
Windows NT4noworks only as viewer
Windows MEyesno problems
Windows 2000yesno problems
wine emulatoryesthe last time I've checked ...

I don't have any Windows NT4 available, so making it work under this environment would be quite impossible.

algorithms
If you wand more information about one of these algorithms just click on it. You will find a small explanation and some sample images.

toolinfoused for processing speed
colormap11 implemented colormaps to choose from visual inspection
very fast
linear/area determinationgets min,max,average temperature from a line or areaheat loss determination
very fast
histogramhistogram viewertemperature frequency
very fast
histeqhistogram equalizationimage enhancement, contrast increase
fast
look-up tablerescales the temperatures from an imageimage enhancement
very fast
optimized look-up tableautomatically rescales the temperaturesimage enhancement
very fast
convolution with 5x5 maskscustom, Laplace, Sobel, Gauss, median, bumpmapedge detect, noise reduction, bluring

fast

Nagao filter(not working properly) noise reduction
quite slow
MHRmaximum homogeneous region filternoise elimination, segmentation, edge and near-edge detection
quite slow
segmentation using CSOMKohonen Self-Organizing Map neural networksegmentation, zone extraction, labeling
slow
level slice re-indexes the image with fewer levelssegmentation
very fast


If you have any questions, bug reports, or just want to say 'hi' please send an e-mail to his address: rodanatsubdimensionpointcom.