Deep convolutional neural networks (DCNNs) have shown significant improvements in remote sensing image scene classification for powerful feature representations. However, because of the high variance and volume limitations available datasets, DCNNs are prone to overfit data used their training. To address this problem, paper proposes a novel framework based on deep Siamese network with rotation...

Abstract In the information age, network security has gradually become a potentially huge problem, especially in medical field, it is essential to ensure accuracy and safety of images, patient needs be included with minimal change. Combined log-polar transform (LPT) discrete cosine (DCT), novel robust watermarking algorithm for proposed. It realized lossless embedding into images. process featu...

In this paper we study the global regularity for solution to Dirichlet problem of equation minimal graphs over a convex domain in hyperbolic spaces. We find that depends only on convexity but independent its smoothness. Basing invariance under translation and rotation transforms, construct super-solution problem, by which prove optimal accurate problem.

This paper presents generalizations to the gray scale and rotation invariant texture classification method based on local binary patterns that we have recently introduced. We derive a generalized presentation that allows for realizing a gray scale and rotation invariant LBP operator for any quantization of the angular space and for any spatial resolution, and present a method for combining mult...

This paper presents a theoretically very simple yet efficient multiresolution approach to gray scale and rotation invariant texture classification based on local binary patterns and nonparametric discrimination of sample and prototype distributions. The method is based on recognizing that certain local binary patterns termed ‘uniform’ are fundamental properties of local image texture, and their...

The Radon transform in combination with self-organizing maps is used to build the rotation invariant systems for categorization of visual objects. The first system has one SOM per the Radon transform direction. The outputs from these directional SOMs that represent positions of the winners and related post-synaptic activities, form the input to the final categorizing SOM. Such a network deliver...

This Traditional template matching methods are not appropriate for the situation of large angle rotation between two images in the online detection for industrial production. Aiming at this problem, Fourier transform algorithm was introduced to correct image rotation angle based on its rotatary invariance in timefrequency domain, orienting image under test in the same direction with reference i...

The rotational invariance is violated in the lightcone quantization method when the Fock space is truncated for practical calculations. To what extent the rotation symmetry is broken in the lightcone quantization approach can be quantified by calculating the explicit rotation dependence of the two-body scattering phase shifts. We analyze the scattering phase shifts in a simple scalar field mode...

In this paper we explore the use of orthogonal functions as generators of representative, compact descriptors of image content. In Image Analysis and Pattern Recognition such descriptors are referred to as image features, and there are some useful properties they should possess such as rotation invariance and the capacity to identify different instances of one class of images. We exemplify our ...