Letter: Learning similarity for semantic images classification

Authors:
Dianhui Wang;Joon Shik Lim;Myung-Mook Han;Byung-Wook Lee
Affiliations:
Department of Computer Science and Computer Engineering, La Trobe University, Melbourne, VIC 3086, Australia and Software College, Kyungwon University, Seongnam, Gyeonggi-Do, 405-760, South Korea;Software College, Kyungwon University, Seongnam, Gyeonggi-Do, 405-760, South Korea;Software College, Kyungwon University, Seongnam, Gyeonggi-Do, 405-760, South Korea;Software College, Kyungwon University, Seongnam, Gyeonggi-Do, 405-760, South Korea
Venue:
Neurocomputing
Year:
2005

Citing 4
Cited 5

Original Contribution: A scaled conjugate gradient algorithm for fast supervised learning

Neural Networks
Similarity Measures

IEEE Transactions on Pattern Analysis and Machine Intelligence
Learning from Data: Concepts, Theory, and Methods

Learning from Data: Concepts, Theory, and Methods
Similarity Searching in Medical Image Databases

IEEE Transactions on Knowledge and Data Engineering

Improved generalization of neural classifiers with enforced internal representation

Neurocomputing
Semantic analysis of real-world images using support vector machine

Expert Systems with Applications: An International Journal
Kernel-based metric learning for semi-supervised clustering

Neurocomputing
A decision support system based on the semantic analysis of melanoma images using multi-elitist PSO and SVM

MLDM'11 Proceedings of the 7th international conference on Machine learning and data mining in pattern recognition
Transductive cost-sensitive lung cancer image classification

Applied Intelligence

Quantified Score

Hi-index	0.01

Visualization

Abstract

While people compare images using semantic concepts, computers compare images using low-level visual features that sometimes have little to do with these semantics. To reduce the gap between the high-level semantics of visual objects and the low-level features extracted from them, in this paper we develop a framework of learning similarity (LS) using neural networks for semantic image classification, where a LS-based k-nearest neighbors (k-NN"L) classifier is employed to assign a label to an unknown image according to the majority of k most similar features. Experimental results on an image database show that the k-NN"L classifier outperforms the Euclidean distance-based k-NN (k-NN"E) classifier and back-propagation network classifiers (BPNC).