This paper describes a procedure developed at University College London for the automatic stereo matching of SAR imagery from NASA's Seasat satellite. The method employed uses Gruen's least squares correlation technique to improve the match accuracy of randomly generated points as they are cascaded down an image pyramid, coupled with a sheet growing mechanism in order to produce a dense array o...

Kernels of a so-called α-scale space ( 12 < α < 1) have the undesirable property that there is no closed-form representation in the spatial domain, despite their simple closedform expression in the Fourier domain. This obstructs a spatial convolution or recursive implementation. For this reason an approximation of the 2D α-kernel in the spatial domain is presented using the well known Gaussian ...

Properties of the receptive fields of simple cells in macaque cortex were compared with properties of independent component filters generated by independent component analysis (ICA) on a large set of natural images. Histograms of spatial frequency bandwidth, orientation tuning bandwidth, aspect ratio and length of the receptive fields match well. This indicates that simple cells are well tuned ...

This insert describes the module akdensity. akdensity extends the official kdensity that estimates density functions by the kernel method. The extensions are of two types: akdensity allows the use of an “adaptive kernel” approach with varying, rather than fixed, bandwidths; and akdensity estimates pointwise variability bands around the estimated density functions.

Detection of objects is extremely important in various aerial vision-based applications. Over the last few years, methods based on convolution neural networks (CNNs) have made substantial progress. However, because large variety object scales, densities, and arbitrary orientations, current detectors struggle with extraction semantically strong features for small-scale by a predefined kernel. To...

In this paper a state of the art sparse coding algorithm for image classification, namely Hierachical Matching Pursuit(HMP), is compared to state of the art algorithms using kernel methods(Efficient Match Kernels, Kernel Descriptors and Hierarchical Kernel Descriptors). HMP is faster and achieves slightly better results than the other algorithms when run over several test-sets. But on the downs...

The paper investigates the application of a recently introduced learning technique, called the relevance vector machine (RVM) to construct a block-adaptive kernel-based nonlinear multiuser detector (MUD) for direct-sequence code-division multiple-access (DS-CDMA) signals transmitted through multipath channels. It is demonstrated that the RVM MUD can closely match the performance of the optimal ...

Remote sensing object detection is a difficult task because it often requires real-time feedback through numerous objects in complex environments. In detection, Feature Pyramids Networks (FPN) have been widely used for better representations based on multi-scale problem. However, the multiple level features cause detectors’ structures to be and makes redundant calculations that slow down detect...