Convolution tree kernels are an efficient and effective method for comparing syntactic structures in NLP methods. However, current kernel methods such as subset tree kernel and partial tree kernel understate the similarity of very similar tree structures. Although soft-matching approaches can improve the similarity scores, they are corpusdependent and match relaxations may be task-specific. We ...

Validation activities for the Medium Resolution Imaging Spectrometer (MERIS) primary radiometric products were carried out using data collected at the Acqua Alta Oceanographic Tower (AAOT) in the northern Adriatic Sea. The intercomparison between the normalized water leaving radiances LWN computed from atmospherically corrected MERIS data and above water radiometric measurements taken with the ...

Object classification and localization are important components of image understanding. For a computer to interact with our world, it will need to identify the objects in our world. At a more basic level, these tasks are crucial to many practical applications: image organization, visual search, autonomous vehicles, and surveillance. This thesis presents alternatives to the currently popular app...

Recently, the sparse coding based codebook learning and local feature encoding have been widely used for image classification. The sparse coding model actually assumes the reconstruction error follows Gaussian or Laplacian distribution, which may not be accurate enough. Besides, the ignorance of spatial information during local feature encoding process also hinders the final image classificatio...

Very high spatial resolution images are now available (about 1m), and processes such as classification or segmentation are less efficient on this kind of images. This paper proposes a method for pre-processing images, which smoothes the heterogeneous areas but perserves the borders of main objects. This process is done by 1. a multi-resolution nonlinear decomposition with a morphological pyrami...

Multiple Kernel Learning, or MKL, extends (kernelized) SVM by attempting to learn not only a classifier/regressor but also the best kernel for the training task, usually from a combination of existing kernel functions. Most MKL methods seek the combined kernel that performs best over every training example, sacrificing performance in some areas to seek a global optimum. Localized kernel learnin...

The Choices operating system splits the microkernel into a machine-independent part and a machine-dependent sub-microkernel. The sub-microkernel, called the nano-kernel in Choices, encapsulates the hardware and presents an idealized machine architecture to the rest of the system. Higher levels of the system access the nano-kernel through a single interface. Nano-kernels are useful because they ...

Multiple kernel learning (MKL) aims at learning a combination of different kernels, instead of using a single fixed kernel, in order to better match the underlying problem. In this paper, we propose the Empirical Optimal Kernel for convex combination MKL. The Empirical Optimal Kernel is based on the theory of kernel polarization, and is the one with the best generalization ability which can be ...

Road extraction in remote sensing images is of great importance for a wide range applications. Because the complex background, and high density, most existing methods fail to accurately extract road network that appears correct complete. Moreover, they suffer from either insufficient training data or costs manual annotation. To address these problems, we introduce new model apply structured dom...