Land-cover classification of partly missing data using support vector machines

Salberg, Arnt-Børre; Jenssen, Robert

doi:10.1080/01431161.2011.648378

Cited by 17 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to train a model that accounts for values in the dataset that are not observed, one has to rely on assumptions that describe how missing data occurs. In this section, we describe the three main missing data mechanisms that characterize the structure of r [17].…”

Section: Missing Data Mechanismsmentioning

confidence: 99%

“…In this experiment, we cluster pixels in high resolution land cover images contaminated with clouds, also used for classification in [16,17]. The data consists of three Landsat ETM+ images covering Hardangervidda in southern Norway, in addition to elevation and slope information.…”

Section: Land Cover Clusteringmentioning

confidence: 99%

“…In the supervised setting, we have e.g. a max-margin framework, where geometric interpretations of the margin is used to account for missing data [1], an approach based on training one SVM per missingness pattern [17] and the "best" Bayesian classifier [12] approach. In the unsupervised setting, there are mixture model formulations accounting for missing features, including both non-Bayesian approaches [5,10] and Bayesian approaches [11].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Spectral Clustering Using PCKID – A Probabilistic Cluster Kernel for Incomplete Data

et al. 2017

Self Cite

View full text Add to dashboard Cite

Abstract. In this paper, we propose PCKID, a novel, robust, kernel function for spectral clustering, specifically designed to handle incomplete data. By combining posterior distributions of Gaussian Mixture Models for incomplete data on different scales, we are able to learn a kernel for incomplete data that does not depend on any critical hyperparameters, unlike the commonly used RBF kernel. To evaluate our method, we perform experiments on two real datasets. PCKID outperforms the baseline methods for all fractions of missing values and in some cases outperforms the baseline methods with up to 25 percentage points.

show abstract

Section: Missing Data Mechanismsmentioning

confidence: 99%

Section: Land Cover Clusteringmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spectral Clustering Using PCKID – A Probabilistic Cluster Kernel for Incomplete Data

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, imputation may often produce suboptimal results, since the assumption of missingness at random [22] may not hold when missingness appears due to the properties of the sensors. Other works that utilize the missingness directly as part of the machine learning approach include Salberg and Jenssen [26], who propose to train SVMs for all types of missingness patterns for landcover classification, and Aksoy et al [23], who present a decision tree based model to data fusion when some modalities are partly missing.…”

Section: Introductionmentioning

confidence: 99%

Urban Land Cover Classification With Missing Data Modalities Using Deep Convolutional Neural Networks

Kampffmeyer

Salberg

Jenssen

2018

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

Automatic urban land cover classification is a fundamental problem in remote sensing, e.g. for environmental monitoring. The problem is highly challenging, as classes generally have high inter-class and low intra-class variance. Techniques to improve urban land cover classification performance in remote sensing include fusion of data from different sensors with different data modalities. However, such techniques require all modalities to be available to the classifier in the decision-making process, i.e. at test time, as well as in training. If a data modality is missing at test time, current state-of-the-art approaches have in general no procedure available for exploiting information from these modalities. This represents a waste of potentially useful information. We propose as a remedy a convolutional neural network (CNN) architecture for urban land cover classification which is able to embed all available training modalities in a so-called hallucination network. The network will in effect replace missing data modalities in the test phase, enabling fusion capabilities even when data modalities are missing in testing. We demonstrate the method using two datasets consisting of optical and digital surface model (DSM) images. We simulate missing modalities by assuming that DSM images are missing during testing. Our method outperforms both standard CNNs trained only on optical images as well as an ensemble of two standard CNNs. We further evaluate the potential of our method to handle situations where only some DSM images are missing during testing. Overall, we show that we can clearly exploit training time information of the missing modality during testing.

show abstract

“…In developing active learning methods, it is desirable that a recognition model has high accuracy while being capable of ranking unseen samples so that those with high ranks can be selected for manual labelling. Among various recognition methods, support vector machines (SVMs) have shown superior performance and have been widely used for VHR images analysis [31], [32], [33], [34]. Nevertheless, common SVMs can not provide accurate ranking information.…”

Section: Introductionmentioning

confidence: 99%

VHR Object Detection Based on Structural Feature Extraction and Query Expansion

Bai

Zhang

Zhou

2014

IEEE Trans. Geosci. Remote Sensing

109

View full text Add to dashboard Cite

Object detection is an important task in very high resolution remote sensing image analysis. Traditional detection approaches are often not sufficiently robust in dealing with the variations of targets, and sometimes suffer from limited training samples. In this paper, we tackle these two problems by proposing a novel method for object detection based on structural feature description and query expansion. The feature description combines both local and global information of objects. After initial feature extraction from a query image and representative samples, these descriptors are updated through an augmentation process to better describe the object of interest. The object detection step is implemented using a ranking support vector machine (SVM), which converts the detection task to a ranking query task. The ranking SVM is firstly trained on a small subset of training data with samples automatically ranked based on similarities to the query image. Then a novel query expansion method is introduced to update the initial object model by active learning with human inputs on ranking of image pairs. Once the query expansion process is completed, which is determined by measuring entropy changes, the model is then applied to the whole target dataset in which objects in different classes shall be detected. We evaluate the proposed method on high resolution satellite images, and demonstrate its clear advantages over several other object detection methods.

show abstract

Land-cover classification of partly missing data using support vector machines

Cited by 17 publications

References 16 publications

Spectral Clustering Using PCKID – A Probabilistic Cluster Kernel for Incomplete Data

Spectral Clustering Using PCKID – A Probabilistic Cluster Kernel for Incomplete Data

Urban Land Cover Classification With Missing Data Modalities Using Deep Convolutional Neural Networks

VHR Object Detection Based on Structural Feature Extraction and Query Expansion

Contact Info

Product

Resources

About