Abstract
This paper presents an image classification method based on a neural network model dealing with tree structures of attentive objects. Apart from regions provided by image segmentation, attentive objects, which are extracted from a segmented image by an attention-driven image interpretation algorithm, are used to construct the tree structure to represent an image. Three combinations of tree structures are investigated, including "image + attentive-object + segments", "image + attentive-objects", as well as "image + segments". Structure based neural networks are trained to classify the images by using the Back Propagation Through Structure (BPTS) algorithm. Experimental results show that the "image + attentive objects" structure is more favorable, comparing with both the other two structures proposed by us and a start-of-art tree structure reported in the literature, in terms of classification rate and computational time.
Original language | English |
---|---|
Title of host publication | 2009 International Joint Conference on Neural Networks, IJCNN 2009 |
Pages | 898-902 |
Number of pages | 5 |
DOIs | |
Publication status | Published - 18 Nov 2009 |
Event | 2009 International Joint Conference on Neural Networks, IJCNN 2009 - Atlanta, GA, United States Duration: 14 Jun 2009 → 19 Jun 2009 |
Conference
Conference | 2009 International Joint Conference on Neural Networks, IJCNN 2009 |
---|---|
Country/Territory | United States |
City | Atlanta, GA |
Period | 14/06/09 → 19/06/09 |
ASJC Scopus subject areas
- Software
- Artificial Intelligence