DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Virtual support vector machines with self-learning strategy for classification of multispectral remote sensing imagery

Geiß, Christian and Aravena Pelizari, Patrick and Blickensdörfer, Lukas and Taubenböck, Hannes (2019) Virtual support vector machines with self-learning strategy for classification of multispectral remote sensing imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 151, pp. 42-58. Elsevier. doi: 10.1016/j.isprsjprs.2019.03.001. ISSN 0924-2716.

[img] PDF - Preprint version (submitted draft)

Official URL: https://www.sciencedirect.com/science/article/abs/pii/S0924271619300620


We follow the idea of learning invariant decision functions for remote sensing image classification with Support Vector Machines (SVM). To do so we generate artificially transformed samples (i.e., virtual samples) from available prior knowledge. Labeled samples closest to the separating hyperplane with maximum margin (i.e., the Support Vectors) are identified by learning an initial SVM model. The Support Vectors are used for generating virtual samples by perturbing the features to which the model should be invariant. Subsequently, the model is relearned using the Support Vectors and the virtual samples to eventually alter the hyperplane with maximum margin and enhance generalization capabilities of decisions functions. In contrast to existing approaches, we establish a self-learning procedure to ultimately prune non-informative virtual samples from a possibly arbitrary invariance generation process to allow for robust and sparse model solutions. The self-learning strategy jointly considers a similarity and margin sampling constraint. In addition, we innovatively explore the invariance generation process in the context of an object-based image analysis framework. Image elements (i.e., pixels) are aggregated to image objects (as represented by segments/superpixels) with a segmentation algorithm. From an initial singular segmentation level, invariances are encoded by varying hyperparameters of the segmentation algorithm in terms of scale and shape. Experimental results are obtained from two very high spatial resolution multispectral data sets acquired over the city of Cologne, Germany, and the Hagadera Refugee Camp, Kenya. Comparative model accuracy evaluations underline the favorable performance properties of the proposed methods especially in settings with very few labeled samples.

Item URL in elib:https://elib.dlr.de/128299/
Document Type:Article
Title:Virtual support vector machines with self-learning strategy for classification of multispectral remote sensing imagery
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Geiß, ChristianUNSPECIFIEDhttps://orcid.org/0000-0002-7961-8553UNSPECIFIED
Aravena Pelizari, PatrickUNSPECIFIEDhttps://orcid.org/0000-0003-0984-4675UNSPECIFIED
Taubenböck, HannesUNSPECIFIEDhttps://orcid.org/0000-0003-4360-9126UNSPECIFIED
Journal or Publication Title:ISPRS Journal of Photogrammetry and Remote Sensing
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:Yes
Page Range:pp. 42-58
Keywords:ClassificationSupport Vector MachinesSelf-learningActive learning heuristicsVery high spatial resolution imagery
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Remote Sensing and Geo Research
Location: Oberpfaffenhofen
Institutes and Institutions:German Remote Sensing Data Center > Geo Risks and Civil Security
Deposited By: Geiß, Christian
Deposited On:05 Aug 2019 10:19
Last Modified:20 Nov 2023 13:55

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.