• Increase font size
  • Default font size
  • Decrease font size

InfoPACK Version 2

E-mail Print PDF

SAR data on its own conveys comparatively little information. We specialise in deriving Knowledge from SAR data by applying a variety of physical models to the data and using rigorous exploitation methods to derive the optimum Information from that data.

In the example above the first figure shows a false colour polarisation image. The second image shows what can be achieved by using an optimised polarimetric segmentation algorithm

InfoPACK provides advanced SAR image understanding algorithms together with an easy to use GUI, image viewer and versatile scripting language, making applications development easy and fast. Particular features of interest include segmentation, classification (supervised and unsupervised), speckle reduction and a host of techniques for data fusion.Central to InfoPACK is a high level library of routines for SAR image understanding, implementing specially developed algorithms optimised for SAR data. The routines optimise memory useage and can handle arbitrarily large images with the Professional Pack. InfoPACK routines may be called from InfoPACK scripts, via the InfoPACK GUI or included in other applications.

Full documentation for InfoPACK is available on-line. You need to be registered to access this.

Case Studies can be found here

Routines within InfoPACK include:

  • Intensity Segmentation (single image; multitemporal; multipolarisation)
  • Speckle Reduction
  • Texture Segmentation
  • Segmentation Postprocessing
  • Classification
  • Edge detection (ab initio and from a prior segmentation)
  • Point Target detection (CFAR-based)
  • Large Area Change detection

Also included are a variety of low-level filters and functions for image manipulation and arithmetic operations.

InfoPACK scripts are based on Python and Numerical Python. This high level language allows for easy and rapid development of applications. The core of the main routines are implemented in C for efficiency.