@ARTICLE {ceamanos*10: a classifier, author={Ceamanos,X. and Waske,B. and Benediktsson,J.A. and Chanussot,J. and Fauvel,M. and Sveinsson, J.R.}, title={{A classifier ensemble based on fusion of support vector machines for classifying hyperspectral data}}, journal={International Journal of Image and Data Fusion}, number={4}, pages={293-307}, year={2010} }

@ARTICLE{dalla mura*10:extended, author={Dalla Mura,M. and Benediktsson,J.A. and Waske,B. and Bruzzone,L.}, title={{Extended profiles with morphological attribute filters for the analysis of hyperspectral data}}, journal={International Journal of Remote Sensing}, number={22}, pages={5975-5991}, year={2010} }

Morphological attribute profiles (APs) are defined as a generalization of the recently proposed morphological profiles (MPs). APs provide a multilevel characterization of an image created by the sequential application of morphological attribute filters that can be used to model different kinds of the structural information. According to the type of the attributes considered in the morphological attribute transformation, different parametric features can be modeled. The generation of APs, thanks to an efficient implementation, strongly reduces the computational load required for the computation of conventional MPs. Moreover, the characterization of the image with different attributes leads to a more complete description of the scene and to a more accurate modeling of the spatial information than with the use of conventional morphological filters based on a predefined structuring element. Here, the features extracted by the proposed operators were used for the classification of two very high resolution panchromatic images acquired by Quickbird on the city of Trento, Italy. The experimental analysis proved the usefulness of APs in modeling the spatial information present in the images. The classification maps obtained by considering different APs result in a better description of the scene (both in terms of thematic and geometric accuracy) than those obtained with an MP.

@ARTICLE{dalla mura*10:morphological, author={Dalla Mura,M. and Benediktsson,J.A. and Waske,B. and Bruzzone,L.}, title={{ Morphological Attribute Profiles for the Analysis of Very High Resolution Images }}, journal={IEEE Transactions on Geoscience and Remote Sensing}, volume={48}, number={10}, pages={3747-3762}, year={2010}, }

The accuracy of supervised land cover classifications depends on factors such as the chosen classification algorithm, adequate training data, the input data characteristics, and the selection of features. Hyperspectral imaging provides more detailed spectral and spatial information on the land cover than other remote sensing resources. Over the past ten years, traditional and formerly widely accepted statistical classification methods have been superseded by more recent machine learning algorithms, e.g., support vector machines (SVMs), or by multiple classifier systems (MCS). This can be explained by limitations of statistical approaches with regard to high-dimensional data, multimodal classes, and often limited availability of training data. In the presented study, MCSs based on SVM and random feature selection (RFS) are applied to explore the potential of a synergetic use of the two concepts. We investigated how the number of selected features and the size of the MCS influence classification accuracy using two hyperspectral data sets, from different environmental settings. In addition, experiments were conducted with a varying number of training samples. Accuracies are compared with regular SVM and random forests. Experimental results clearly demonstrate that the generation of an SVM-based classifier system with RFS significantly improves overall classification accuracy as well as producer’s and user’s accuracies. In addition, the ensemble strategy results in smoother, i.e., more realistic, classification maps than those from stand-alone SVM. Findings from the experiments were successfully

@ARTICLE{5439692, author={Waske, B. and van der Linden, S. and Benediktsson, J. A. and Rabe, A. and Hostert, P.}, journal={Geoscience and Remote Sensing, IEEE Transactions on}, title={Sensitivity of Support Vector Machines to Random Feature Selection in Classification of Hyperspectral Data}, year={2010}, month={july }, volume={48}, number={7}, pages={2880 -2889}, keywords={}, doi={10.1109/TGRS.2010.2041784}, ISSN={0196-2892},}

Hyperspectral imaging provides detailed spectral and spatial information from the land cover that enables a precise differentiation between various surface materials. On the other hand, the performance of traditional and widely used statistical classification methods is often limited in this context. Thus, alternative methods are required. In the study presented here, the performance of two machine learning techniques – Support Vector Machines (SVM) and Random Forests (RF) – are investigated and the classification results are compared to well known methods (i.e., maximum likelihood classifier and spectral angle mapper). The classifiers are applied to an AVIRIS data set that was acquired near the volcano Hekla in Iceland. The results clearly show the advantages of the two proposed classifier algorithms in terms of accuracy. They significantly outperform the other methods and achieve overall accuracies of approximately 90%. Although SVM and RF show some diversity in the classification results, the global performance of the two classifiers is very similar. Thus, both methods can be considered attractive for the classification of hyperspectral data.

@ARTICLE{waske*09:mapping, author = {Waske, Bj\"{o}rn and Benediktsson, Jon Atli and Árnason, Kolbein and Sveinsson, Johannes R.}, title = {Mapping of hyperspectral AVIRIS data using machine learning algorithms}, journal={Canadian Journal of Remote Sensing}, volume={35}, year = {2009}, number={S1}, pages = {106-116} }

SAR data are almost independent from weather conditions, and thus are well suited for mapping of seasonally changing variables such as land cover. In regard to recent and upcoming missions, multitemporal and multi-frequency approaches become even more attractive. In the present study, classifier ensembles (i.e., boosted decision tree and random forests) are applied to multi-temporal C-band SAR data, from different study sites and years. A detailed accuracy assessment shows that classifier ensembles, in particularly random forests, outperform standard approaches like a single decision tree and a conventional maximum likelihood classifier by more than 10% independently from the site and year. They reach up to almost 84% of overall accuracy in rural areas with large plots. Visual interpretation confirms the statistical accuracy assessment and reveals that also typical random noise is considerably reduced. In addition the results demonstrate that random forests are less sensitive to the number of training samples and perform well even with only a small number. Random forests are computationally highly efficient and are hence considered very well suited for land cover classifications of future multifrequency and multitemporal stacks of SAR imagery.

@article{Waske2009450, title = "Classifier ensembles for land cover mapping using multitemporal SAR imagery", journal = "ISPRS Journal of Photogrammetry and Remote Sensing", volume = "64", number = "5", pages = "450 - 457", year = "2009", note = "Theme Issue: Mapping with SAR: Techniques and Applications", issn = "0924-2716", doi = "DOI: 10.1016/j.isprsjprs.2009.01.003", url = "http://www.sciencedirect.com/science/article/B6VF4-4VP12G6-1/2/fc8ceb15fc4b40b2507ba5482ce204e8", author = "Björn Waske and Matthias Braun", keywords = "Decision tree", keywords = "Random forests", keywords = "Boosting", keywords = "Multitemporal SAR data", keywords = "Land cover classification" }

A strategy for the joint classification of multiple segmentation levels from multisensor imagery is introduced by using synthetic aperture radar and optical data. At first, the two data sets are separately segmented, creating independent aggregation levels at different scales. Each individual level from the two sensors is then preclassified by a support vector machine (SVM). The original outputs of each SVM, i.e., images showing the distances of the pixels to the hyperplane fitted by the SVM, are used in a decision fusion to determine the final classes. The fusion strategy is based on the application of an additional classifier, which is applied on the preclassification results. Both a second SVM and random forests (RF) were tested for the decision fusion. The results are compared with SVM and RF applied to the full data set without preclassification. Both the integration of multilevel information and the use of multisensor imagery increase the overall accuracy. It is shown that the classification of multilevel-multisource data sets with SVM and RF is feasible and does not require a definition of ideal aggregation levels. The proposed decision fusion approach that applies RF to the preclassification outperforms all other approaches

@ARTICLE{4481234, title={Classifying Multilevel Imagery From SAR and Optical Sensors by Decision Fusion}, author={Waske, B. and van der Linden, S.}, journal={Geoscience and Remote Sensing, IEEE Transactions on}, year={2008}, month={May }, volume={46}, number={5}, pages={1457-1466}, doi={10.1109/TGRS.2008.916089}, ISSN={0196-2892}, }

The classification of multisensor data sets, consisting of multitemporal synthetic aperture radar data and optical imagery, is addressed. The concept is based on the decision fusion of different outputs. Each data source is treated separately and classified by a support vector machine (SVM). Instead of fusing the final classification outputs (i.e., land cover classes), the original outputs of each SVM discriminant function are used in the subsequent fusion process. This fusion is performed by another SVM, which is trained on the a priori outputs. In addition, two voting schemes are applied to create the final classification results. The results are compared with well-known parametric and nonparametric classifier methods, i.e., decision trees, the maximum-likelihood classifier, and classifier ensembles. The proposed SVM-based fusion approach outperforms all other approaches and significantly improves the results of a single SVM, which is trained on the whole multisensor data set.

@ARTICLE{4358858, title={Fusion of Support Vector Machines for Classification of Multisensor Data}, author={Waske, B. and Benediktsson, J.A.}, journal={Geoscience and Remote Sensing, IEEE Transactions on}, year={2007}, month={Dec. }, volume={45}, number={12}, pages={3858-3866}, doi={10.1109/TGRS.2007.898446}, ISSN={0196-2892}, }

In the proposed approach, the well-known enhanced Lee filter is modified to allow the integration of feature outlines-previously extracted from segmented optical images. The filter is applied to several ENVISAT ASAR images that cover urban, agricultural, and forest areas during different plant phenological stages. The performance of this segment-based speckle filter is compared to those of other filters using ratio images, visual interpretation, and statistical indexes. The approach reduces the loss of radiometry and spatial information. It performs comparable to more complex methods and outperforms common techniques

@ARTICLE{4156176, title={A Segment-Based Speckle Filter Using Multisensoral Remote Sensing Imagery}, author={Waske, B. and Braun, M. and Menz, G.}, journal={Geoscience and Remote Sensing Letters, IEEE}, year={2007}, month={April }, volume={4}, number={2}, pages={231-235}, doi={10.1109/LGRS.2006.888849}, ISSN={1545-598X},