Dimensionality reduction for content-based image classification

Effective ways of organizing image descriptors is a critical design step of content-based image classification systems. Suitable descriptors are selected according to the problem domain for generating the feature space. Using several descriptors improves accuracy of representation but risen some cha...

Full description

Saved in:
Bibliographic Details
Main Authors: Mrowka, E., Dorado, A., Pedrycz, W., Izquierdo, E.
Format: Conference Proceeding
Language:English
Subjects:
Design engineering
Image classification
Image databases
Independent component analysis
Indexing
Information retrieval
Organizing
Principal component analysis
Spatial databases
Testing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Effective ways of organizing image descriptors is a critical design step of content-based image classification systems. Suitable descriptors are selected according to the problem domain for generating the feature space. Using several descriptors improves accuracy of representation but risen some challenges such as non linear combination, expensive computation and the curse of dimensionality. In This work an approach using a non parametric statistical test for effective dimensionality reduction is presented. The proposed method facilitates feature discrimination and keeps relevant information.
ISSN:1093-9547
DOI:10.1109/IV.2004.1320180