Back to Ari Visa's Homepage





Knowledge Mining and Managing in a Distributed Image Datawarehouse - DIGGER


This project considers large image databases. The problems: how to find information, how to get an idea of the contents of the database, and how to manage a distributed, heterogeneous database, are under considerations. Some new tools based on adaptive agents have been developed for image retrieval. An architecture to manage a distributed datawarehouse has been suggested. PicSom has been tested to image browsing purposes. The process to improve the accuracy and the speed in retrieval from a large image database is now under the work.


The results of this project have been documented in many publications. The publications are listed here. In addition five persons have done their Master Thesis in this project and two persons their Licentiate Thesis:

  • Amelia Gutiérrez Ortega, Experimental Study of Content Based Image Retrieval, 2002
  • Jussi Pakkanen, Sisältöpohjainen haku paperivirhetietokannassa PicSOM-järjestelmän avulla, 2002
  • Eija Salko, Agenttipohjainen tiedonhaku hajautetussa tietokannassa, 2003
  • Joona Petrell, Tilastomatemaattiset uudelleenotantamenetelmät hakujärjestelmien arvioinnissa, 2004
  • Rami Rautkorpi, Shape features in the classification and retrieval of surface defect images,2005
  • Kati Kuusijärvi, Indeksointi kuvatietokannoissa, 2005
  • Iivari Kunttu, Hahmontunnistusmenetelmiä paperikuvatietokannan indeksointiin sisältöpohjaista luokittelua ja hakua varten, 2003
  • Leena Lepistö, Hahmontunnistusmenetelmiä kivikuvien luokitteluun, 2004

Project information

The research work is done in Institute of Signal Processing at Tampere University of Technology and in Laboratory of Computer and Information Science at Helsinki University of Technology, Finland during 1.1.2001 - 31.12.2004. The budget of this 4 years project is roughly 500 000 €. The work is carried out in close co-operation with the following industrial companies:

The project manager is professor Ari Visa and the project is funded by TEKES.

Project group

At this moment the following persons belong to Digger project group:

  • Pekka Yrjölä, Tekes
  • Olli Ventä, VTT
  • Juhani Rauhamaa, ABB Oy
  • Jorma Autio, Saanio & Riekkola Oy
  • Hanna Malmlund, Saanio & Riekkola Oy
  • Jukka Iivarinen, HUT
  • Jussi Pakkanen, HUT
  • Rami Rautkorpi, HUT
  • Ari Visa, TUT
  • Iivari Kunttu, TUT
  • Leena Lepistö, TUT
  • Joona Petrell, TUT
  • Kati Kuusijärvi, TUT

Photo of the
           project group

Part of the project group in a meeting in January 2004.