Hierarchical Multilabel Ship Classification in Remote Sensing Images Using Label Relation Graphs
- 15 September 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892), 1-13
- https://doi.org/10.1109/tgrs.2021.3111117
Abstract
Hierarchical multilabel classification (HMC) assigns multiple labels to each instance with the labels organized under hierarchical relations. In ship classification in remote sensing images, depending on the expert knowledge and image quality, the same type of ships in different remote sensing images may be annotated with different class labels from coarse to fine levels such as merchant ship (MS) or container ship (CTS). In this article, we propose a novel deep network with two output channels and their associated loss functions to learn an HMC classifier using samples labeled at different levels in the hierarchy. In the proposed network, a hierarchy and exclusion (HEX) graph is introduced to model the label hierarchy, which satisfies hierarchical constraints by encoding semantic relations between any two labels. The output nodes of the first channel are organized according to the HEX graph, and its corresponding probabilistic classification loss is built to reflect the hierarchical structure of the HEX graph. On the other hand, the output nodes of the second channel only represent the finest grained (last level in the hierarchy) classes, and its multiclass cross-entropy loss is designed to enhance the discriminative power of the HMC classifier on the last level labels, which is also compatible with constraints in the HEX graph. The combination of these two losses from two output channels can effectively transfer the hierarchical information of ship taxonomy during network training. Experimental results on two commonly used ship datasets demonstrate that the proposed method outperforms the state-of-the-art HMC approaches, and is especially advantageous when trained with fewer fine-grained samples.Keywords
Funding Information
- National Key Research and Development Program of China (2018AAA0100500)
- National Natural Science Foundation of China (62071421)
This publication has 46 references indexed in Scilit:
- Hierarchical multi-label classification using local neural networksJournal of Computer and System Sciences, 2013
- Hierarchical classification of diatom images using ensembles of predictive clustering treesEcological Informatics, 2012
- Hierarchical annotation of medical imagesPattern Recognition, 2011
- Maritime surveillance: Tracking ships inside a dynamic background using a fast level-setExpert Systems with Applications, 2011
- A survey of hierarchical classification across different application domainsData Mining and Knowledge Discovery, 2010
- Predicting gene function using hierarchical multi-label decision tree ensemblesBMC Bioinformatics, 2010
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009
- Decision trees for hierarchical multi-label classificationMachine Learning, 2008
- Hierarchical multi-label prediction of gene functionBioinformatics, 2006
- Automated categorization in the international patent classificationACM SIGIR Forum, 2003