Multiple-relations-constrained image classification with limited training samples via Pareto optimization

It is a significant challenge to classify images using limited training samples. To this end, we formulate image classification as a multi-task multi-view (MTMV) learning problem and propose a novel Pareto optimization-based method to find the solution. Specifically, we first build a multi-objective...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Neural computing & applications Jg. 31; H. 10; S. 6821 - 6842
Hauptverfasser: Zhou, Di, Wang, Jun, Jiang, Bin, Li, Yajun
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London Springer London 01.10.2019
Springer Nature B.V
Schlagworte:
Algorithms
Artificial Intelligence
Classification
Classifiers
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Science
Data Mining and Knowledge Discovery
Image classification
Image Processing and Computer Vision
Multiple objective analysis
Optimization
Original Article
Pareto optimization
Pareto optimum
Probability and Statistics in Computer Science
Training
Multiple-relations-constrained
Multi-objective optimization
Image classification
Multi-task multi-view learning
ISSN:0941-0643, 1433-3058
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:It is a significant challenge to classify images using limited training samples. To this end, we formulate image classification as a multi-task multi-view (MTMV) learning problem and propose a novel Pareto optimization-based method to find the solution. Specifically, we first build a multi-objective multiple-relations-constrained MTMV model called M 4 to formulate the procedure of image classification. This model integrates comprehensive relations so that more knowledge can be used when classifying the images. We formulate the model as a multi-objective optimization problem, which addresses conflicts between the inconsistencies of each item in the model and the limitation on the number of relations. To generate the final classifier for each image classification task, an effective Pareto optimization-based algorithm Pareto-M 4 is proposed. Pareto-M 4 first generates the Pareto-optimal solutions using a novel multi-objective solver MOQPSO and then obtains the final classifiers from all the Pareto-optimal solutions using a recombination procedure. Experiments on various real-world image data sets demonstrate the effectiveness of the proposed image classification method when limited training samples are given.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0941-0643
1433-3058
DOI:10.1007/s00521-018-3491-4