Sitemap Contact Events New Home

Database of research projects FEDRA


Research actions



Research and applications > Projectendatabanken > Database of research projects FEDRA

Data analysis in hyperspectral remote sensing (HYPERCRUNCH)

Research project S0/00/005 (Research action S0)

Persons :

  • Prof. dr.  COPPIN Pol - Katholieke Universiteit Leuven (K.U.Leuven)
    Financed belgian partner
    Duration: 15/12/2001-31/12/2004
  • Prof. dr.  SCHEUNDERS Paul - Universiteit Antwerpen (UA)
    Financed belgian partner
    Duration: 15/12/2001-31/12/2004
  • Dhr.  DEBRUYN Walter - Vlaamse Instelling voor Technologisch Onderzoek (VITO)
    Financed belgian partner
    Duration: 15/12/2001-31/12/2004

Description :

Topic and context

The project lies within the scope of the "Advanced generic data processing and innovation" research field.

Image spectroscopy (IS) is a remote detection technique whose popularity is steadily increasing, not just internationally, but also within the framework of ‘Technology & Application‘ developments in Belgium.

IS’s specific high spectral and spatial resolution has one major drawback: image spectrometers supply massive quantities of measurement data. This makes selection of a limited number of relevant bands without any information loss for a given application a critical point for every IS application and an initial research topic for this project. A second topic involves the development of algorithms for improving classification in relation to spectral separation and in relation to the classification procedures themselves. The mathematical algorithms will therefore be developed as independently as possible from applications so as to enable their automation and implementation in operational data processing chains, such as those for the APEX sensor. A better-quality end product will also be made available to the scientific community.


The general objective of the proposed project is to improve data mining from hyperspectral data cubes, the aim of which is two-fold: to serve the scientific community more effectively; and extend image-processing chains to this specific type of data.

The clearest possible illustration of objectives is provided by the background and role of the labs participating in the project structure.

Two partners, VITO and RUCA, provide the consortium with complementary image-processing expertise. VITO is contributing its experiences from the field of IS image processing and data cube correction and final processing, plus its knowledge of "conventional" data mining from hyperspectral datasets. RUCA, as an internationally acknowledged research centre for the development of image interpretation and processing in (for example) the biomedical sector, provides the project with its expertise in terms of the development of image-processing algorithms, including the required theoretical mathematical and physical background. RUCA and VITO will translate the developed data-mining algorithms into functional coding and VITO will incorporate them into operational data-processing chains. VITO has the expertise that is needed in order to manage these types of operational data-processing chain, e.g. those intended for the APEX sensor.

Quite clearly, the data-reduction techniques and mathematical algorithms expected to emerge here as end products from the project will be tested and validated. The application, which this proposal is considering to this end, is precision farming, or more specifically the detection of stress within orchards. Case studies will be conducted on test plots at the Royal Research Station in Gorsem. Validation is the responsibility of KU Leuven, the consortium’s third partner. KU Leuven is providing its expertise in the interpretation of quantitative information regarding vegetation processes and parameters, from leaves through to complex spatial scales. Ultimately, the complementarity between the different research groups will leave the consortium in a position to link the spectral information to physical reality.

Documentation :

About this website


© 2018 Science Policy PPS