A relative evaluation of multiclass image classification by support vector machines

By Giles M. Foody, Ajay Mathur. IEEE Transactions on Geoscience and Remote Sensing, 2004

1,088 citations recorded by the source corpus.

Support vector machines (SVMs) have considerable potential as classifiers of remotely sensed data. A constraint on their application in remote sensing has been their binary nature, requiring multiclass classifications to be based upon a large number of binary analyses. Here, an approach for multiclass classification of airborne sensor data by a single SVM analysis is evaluated against a series of classifiers that are widely used in remote sensing, with particular regard to the effect of training set size on classification accuracy. In addition to the SVM, the same datasets were classified using a discriminant analysis, decision tree, and multilayer perceptron neural network. The accuracy statements of the classifications derived from the different classifiers were compared in a statistically rigorous fashion that accommodated for the related nature of the samples used in the analyses. For each classification technique, accuracy was positively related with the size of the training set. In general, the most accurate classifications were derived from the SVM approach, and with the largest training set the SVM classification was significantly (p < 0.05)more accurate (93.75%) than that derived from the discriminant analysis (90.00%) and decision tree algorithms (90.31%). Although each classifier could yield a very accurate classification, > 90% correct, the classifiers differed in the ability to correctly label individual cases and so may be suitable candidates for an ensemble-based approach to classification.

Subfields: Computer Vision and Pattern Recognition, Ecology, Media Technology

View Paper: https://www.semanticscholar.org/paper/8a59973ec0fa70a82f512ea2edcf890129ea17f7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Ecology

In Media Technology

Citations in the corpus

Spectral and Spatial Classification of Hyperspectral Data Using SVMs and Morphological Profiles (2008, result): “Multiclass SVM performances were also positively compared with a discriminant analysis classiﬁer, a decision tree classiﬁer and a feedforward neural network classiﬁer with a limited training set [25].”
Geological mapping using remote sensing data: A comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information (2014, methodology): “…on the use of MLAs for the supervised classification of remote sensing data has been for the prediction of land cover or vegetation classes (e.g., Foody and Mathur, 2004; Ham et al., 2005; Huang et al., 2002; Pal, 2005; Song Contents lists available at ScienceDirect journal homepage:…”
Feature Selection for Classification of Hyperspectral Data by SVM (2010, background): “A detailed discussion of the computational aspects of SVM can be found in [14] and [45], with many examples also in the remote sensing literature [19], [21], [46], [47].”
Deep Learning-Based Classification of Hyperspectral Data (2014, methodology): “Traditional hyperspectral data classification methods use spectral information only, and the classification algorithms typically include parallelepiped classification, k-nearest-neighbors, maximumlikelihood, minimum distance, and logistic regression [13].”
An SVM Ensemble Approach Combining Spectral, Structural, and Semantic Features for the Classification of High-Resolution Remotely Sensed Imagery (2013, methodology): “In this paper, the one-versus-all approach [48] is used for the multiclass SVM soft output.”
Integrating remotely sensed leaf area index and leaf nitrogen accumulation with RiceGrow model based on particle swarm optimization algorithm for rice grain yield assessment (2014, methodology): “By combining a support vector machine method with field surveys and ground-based spectral data, the supervised classification and reflectance threshold segmentation of HJ-1A/B images were carried out to extract the rice planting area.(22,23) Finally, an Wang et al.”
Invasion of glossy privet (Ligustrum lucidum) and native forest loss in the Sierras Chicas of Córdoba, Argentina (2010, methodology): “Thus, rather than describing statistical characteristics of classes, SVMs use only training samples that describe class boundaries (Foody and Mathur 2004).”
Hyperspectral Image Classification With Independent Component Discriminant Analysis (2011, neutral): “Since preliminary experiments have shown that the smallest and biggest values of ICs are not useful, because they do not contain enough information or they have redundant features, to avoid a large computational time, the range investigated was [10–30] in the case of ROSIS and AVIRIS Hekla data sets”
Deep Convolutional Neural Networks for Hyperspectral Image Classification (2015, background): “Taking advantages of the rich spectral information, numerous traditional classification methods, such as k-nearest-neighbors (k-nn), minimum distance, and logistic regression [2], have been developed.”