Laboratory of Computer and Information Science
Project Home Page


Contact Info

Publications

Local Info












VISUAL FAULT ANALYSIS OF RUNNING PAPER WEB PROJECT

The Visual Fault Analysis of Running Paper Web project in the Laboratory of Computer and Information Science at Helsinki University of Technology belongs to a larger research project IMPRESS - Intelligent Methods for Processing and Exploration of Signals and Systems. IMPRESS is part of the TEKES's technology programme Adaptive and Intelligent Systems Applications (AISA). The project has been running since 1994 and it ends in the end of 1999.

The industrial partner is ABB Industry Oy and the research partner (in 1995-1997) the Signal Processing Laboratory at Tampere University of Technology.

The goal of the project is to develop a classification system for fast, real-time web surface inspection that learns texture and shape characteristics of target objects from examples without (almost) any a priori information on the target objects.

High speed requirements of a web inspection system limit the number of techniques that can be applied. Historically defect detection of web surfaces has been accomplished by hardware solutions for thresholding and matched filters. These techniques have made possible to detect only the most basic defect types. Detection of more complicated but critical defect types has remained unreliable. In this project more sophisticated methods have been developed. Use of texture and more complicated classifiers have become possible due to new sensor technology, increased calculation capability of computers and specialized hardware.




This page is maintained by jukka.iivarinen@hut.fi.
Last modified 1.3.1999.
URL: http://www.cis.hut.fi/research/IA/paper/