Skip to Main content Skip to Navigation

Analyse de scènes naturelles par Composantes Indépendantes

Abstract : In this thesis, Independent Component Analysis (ICA) is used to compute features extracted from natural images. The use of ICA is justified in the context of classification of natural images for two reasons. On the one hand the model of image suggests that the underlying statistical principles may be the same as those that determine the structure of the visual cortex. As a consequence, the filters that ICA produces are adapted to the statistics of natural images. On the other hand, we adopt a non-parametric approach that requires density estimation in many dimensions, and independence between features appears as a solution to overthrow the “curse of dimensionality”. In parallel, we have conducted a psychophysical experiment to determine a human perception space in which we have identified perceptive categories. These categories and the distances between images are emphasised by analysing the human response similarities with a non-linear multidimensional scaling technique. Analysis of the results allows us to appreciate the influence of colour and put into relief perceptive asymmetries.
Complete list of metadata

Cited literature [254 references]  Display  Hide  Download
Contributor : Hervé Le Borgne <>
Submitted on : Friday, April 16, 2004 - 5:02:52 PM
Last modification on : Friday, March 19, 2021 - 5:00:28 PM
Long-term archiving on: : Wednesday, November 23, 2016 - 4:18:47 PM


  • HAL Id : tel-00005925, version 1



Hervé Le Borgne. Analyse de scènes naturelles par Composantes Indépendantes. Interface homme-machine [cs.HC]. Institut National Polytechnique de Grenoble - INPG, 2004. Français. ⟨tel-00005925⟩



Record views


Files downloads