Gustavo Camps-Valls

Toolbox		Info
	SIMIT	SIMIT: A simple Matlab toolbox for information theory quantity estimation Entropies, negentropies, mutual information, and divergences
	SIMFEAT	SIMFEAT: A simple Matlab toolbox of linear and kernel feature extraction methods - Linear methods: PCA, MNF, PCC, PLS, OPLS, LDA, PLDA, WLDA, RLDA, OLDA, ULDA - Kernel feature extractors: KPCA, KICA, HSIC, KCCA, KPLS, KOPLS, KSNR and KECA
	MSVR	Multi-output Support Vector Regression (M-SVR) Standard SVR formulation only considers the single-output problem. In the case of several output variables, other methods (neural networks, kernel ridge regression) must be deployed, but the good properties of SVR are lost: hinge-loss function and sparsity. The proposed model M-SVR extends the single-output SVR by taking into account the nonlinear relations between features but also among the output variables, which are typically inter-dependent.
	PD-SVR	Profile-dependent (Adaptive) Support Vector Regression. Different samples in a distribution should be penalized differently in a SVR, and also should their confidence be tailored. This basic idea leads to the selection of a different epsilon and C parameters for training the SVR according to the a priori relevance of the sample. This concept was originally applied to pharmacy time-series prediction, regression problems in bioinformatics, remote sensing and image coding.
	e-Huber SVR	Epsilon-Huber Support Vector Regression. The combination of the classical Vapnik's e-insensitive loss function and the Huber cost function leads to enhanced performance when different noise sources are present in the data. This cost function has been applied to system identification, gamma-filtering, and recently to SVR. Attention: the code we provide is specific for the problem in "Robust Support Vector Regression for Biophysical Parameter Estimation from Remotely Sensed Images" Gustavo Camps-Valls, L. Bruzzone, José L. Rojo-Álvarez, Farid Melgani. IEEE Geoscience and Remote Sensing Letters, July 2006. Volume: 3, Issue: 3, pp. 339-343. A much simpler way to implement the loss consists just in adding an extra-regularization to the diagonal kernel matrix (see this paper for details).
	SSKOSP	Semi-supervised Kernel Orthogonal Subspace Projection. The Orthogonal Subspace Projection (OSP) algorithm is substantially a kind of matched filter that requires the evaluation of a prototype for each class to be detected. The kernel OSP (KOSP) has recently demonstrated improved results for target detection in hyperspectral images. We present a semi-supervised graph-based approach to improve KOSP. The proposed algorithm deforms the kernel by approximating the marginal distribution using the unlabeled samples. Two further improvements are presented. First, a contextual selection of unlabeled samples is proposed. This strategy helps in better modeling the data manifold and thus improved sensitivity-specificity rates are obtained. Second, given the high computational burden involved, we present two alternative formulations based on the Nyström method and the Incomplete Cholesky Factorization to achieve operational processing times.
	Bagged SVM	Bagged Support Vector Machine. A semi-supervised SVM is presented for the classification of remote sensing images. The method exploits the wealth of unlabeled samples for regularizing the training kernel representation locally by means of cluster kernels. The method learns a suitable kernel directly from the image, and thus avoids assuming a priori signal relations by using a predefined kernel structure. Good results are obtained in image classification examples when few labeled samples are available. The method scales almost linearly with the number of unlabeled samples and provides out-of-sample predictions.
	SSSVR	Semi-supervised Support Vector Regression. This paper presents two kernel-based methods for semi-supervised regression. The methods rely on building a graph or hypergraph Laplacian with both the available labeled and unlabeled data, which is further used to deform the training kernel matrix. The deformed kernel is then used for support vector regression (SVR). Given the high computational burden involved, we present two alternative formulations based on the Nyström method and the Incomplete Cholesky Factorization to achieve operational processing times. The semi-supervised SVR algorithms are sucessfully tested in multiplatform LAI estimation and oceanic chlorophyll concentration prediction. Experiments are carried out with both multispectral and hyperspectral data, demonstrating good generalization capabilities when low number of labeled samples are available, which is usually the case in biophysical parameter estimation.
	KERVI	KERVI: Kernels in ENVI We are integrating many kernel methods in ENVI (still under development)
	Graph-kernel	Graph kernel for spatio-spectral classification This paper presents a graph kernel for spatio-spectral remote sensing image classification with support vector machines (SVM). The method considers higher order relations in the neighborhood (beyond pairwise spatial relations) to iteratively compute a kernel matrix for SVM learning. The proposed kernel is easy to compute and constitutes a powerful alternative to existing approaches. The capabilities of the method are illustrated in several multi- and hyperspectral remote sensing images acquired over both urban and agricultural areas.
	SSL Graph-based	Semi-supervised Graph-based Classification. A graph-based method for semi-supervised learning: essentially an affinity matrix is computed, the graph Laplacian is normalized, and a spreading function is iterated until convergence. This algorithm can be understood intuitively in terms of spreading activation networks from experimental psychology, and explained as random walks on graphs. We successfully apply it to hyperspectral image classification. It incorporates contextual information through a full family of composite kernels. Noting that the graph method relies on inverting a huge kernel matrix formed by both labeled and unlabeled samples, we originally introduce the Nyström method in the formulation to speed up the classification process.
	KARMA	Kernel AutoRegressive Moving Average with the Support Vector Machine. Nonlinear system identification based on Support Vector Machines (SVM) has been usually addressed by means of the standard SVM regression (SVR), which can be seen as an implicit nonlinear Auto-Regressive and Moving Average (ARMA) model in some Reproducing Kernel Hilbert Spaces (RKHS). The proposal here is twofold: First, the explicit consideration of an ARMA model in RKHS (SVM-ARMA2K) is originally proposed. Second, a general class of SVM-based system identification nonlinear models is presented, based on the use of composite Mercer’s kernels.
	SVMTools	The standard libsvm with some add-ons: Precomputed kernels, e-Huber cost function, accuracy assessment, and other funny things for kernel methods.
	GPCA/RBIG	Gaussianization based on PCA The GPCA Toolbox is a Matlab Toolbox for estimates the Gaussianization transformation based on PCA (or any kind of orthogonal transform, random rotations included!) from given multidimensional signals.
	VistaCoderTools	VistaCodeTools. The Image Coding Toolbox is a Matlab Toolbox for achromatic and color image coding that includes a set of DCT algorithms based on Human Vision Models of different accuracy and SVM selection of transform coefficients.
	VistaRestoreTools	VistaRestoreTools. The Image Restoration Toolbox is a Matlab Toolbox for image restoration that includes (1) classical regularization techniques, (2) classical wavelet thresholding techniques, (3) regularization functionals based on non-linear human vision models, and (4) denoising techniques based on Kernel regression in wavelet domains.
	VistaQualityTools	VistaQualityTools. The Image Quality Toolbox is a Matlab Toolbox for full reference color (and also achromatic) image quality assessment based on divisive normalization models in DCT and wavelet domains.

Sections

Personal tools