Development of a Planar Coupler with Reduced Dimensions Using Idle Stubs

Author(s):  
Dilshod C. Ravshanov ◽  
Abdulmusavvir A. Karimov ◽  
Evgeniy B. Bablyuk
Keyword(s):  
2021 ◽  
Vol 11 (11) ◽  
pp. 5235
Author(s):  
Nikita Andriyanov

The article is devoted to the study of convolutional neural network inference in the task of image processing under the influence of visual attacks. Attacks of four different types were considered: simple, involving the addition of white Gaussian noise, impulse action on one pixel of an image, and attacks that change brightness values within a rectangular area. MNIST and Kaggle dogs vs. cats datasets were chosen. Recognition characteristics were obtained for the accuracy, depending on the number of images subjected to attacks and the types of attacks used in the training. The study was based on well-known convolutional neural network architectures used in pattern recognition tasks, such as VGG-16 and Inception_v3. The dependencies of the recognition accuracy on the parameters of visual attacks were obtained. Original methods were proposed to prevent visual attacks. Such methods are based on the selection of “incomprehensible” classes for the recognizer, and their subsequent correction based on neural network inference with reduced image sizes. As a result of applying these methods, gains in the accuracy metric by a factor of 1.3 were obtained after iteration by discarding incomprehensible images, and reducing the amount of uncertainty by 4–5% after iteration by applying the integration of the results of image analyses in reduced dimensions.


Mathematics ◽  
2021 ◽  
Vol 9 (11) ◽  
pp. 1298
Author(s):  
Mitzi Cubilla-Montilla ◽  
Ana Belén Nieto-Librero ◽  
M. Purificación Galindo-Villardón ◽  
Carlos A. Torres-Cubilla

The HJ biplot is a multivariate analysis technique that allows us to represent both individuals and variables in a space of reduced dimensions. To adapt this approach to massive datasets, it is necessary to implement new techniques that are capable of reducing the dimensionality of the data and improving interpretation. Because of this, we propose a modern approach to obtaining the HJ biplot called the elastic net HJ biplot, which applies the elastic net penalty to improve the interpretation of the results. It is a novel algorithm in the sense that it is the first attempt within the biplot family in which regularisation methods are used to obtain modified loadings to optimise the results. As a complement to the proposed method, and to give practical support to it, a package has been developed in the R language called SparseBiplots. This package fills a gap that exists in the context of the HJ biplot through penalized techniques since in addition to the elastic net, it also includes the ridge and lasso to obtain the HJ biplot. To complete the study, a practical comparison is made with the standard HJ biplot and the disjoint biplot, and some results common to these methods are analysed.


2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Yuqi Wang ◽  
Soroush Arabi ◽  
Klaus Kern ◽  
Markus Ternes

AbstractSymmetries in nanoscale structures can be decisive for their structural, electronic, and magnetic properties, particularly in systems with reduced dimensions. Here we show that the symmetries of a flat metal-organic molecule adsorbed on a transition metal dichalcogenide, a 2-dimensional layered material, have a dramatic effect on the total spin and the intramolecular spin-spin interactions. Using a scanning probe microscope, we find two different molecular spin states by modifying the symmetry of the molecules via the twist angle to the substrate. Additionally, we observe significant non-collinear Dzyaloshinskii–Moriya interaction between two electron spins on the molecule induced by the spin-orbit coupling of the van der Waals coupled layered material with broken inversion symmetry. Our work opens a path for modifying the spin by exploiting symmetries and for studying the nature of surface-induced non-collinear spin-spin interaction within a single molecule which might allow the realization of more complex topological spin structures.


2008 ◽  
Vol 56 (2) ◽  
pp. 85-96 ◽  
Author(s):  
Frederico de Moraes Rudorff ◽  
Douglas Francisco Marcolino Gherardi

The present work aimed to examine the potentials of SAR RADARSAT-1 images to detect emergent coral reefs at the Environmental Protection Area of "Costa dos Corais". Multi-view filters were applied and tested for speckle noise reduction. A digital unsupervised classification based on image segmentation was performed and the classification accuracy was evaluated by an error matrix built between the SAR image classification and a reference map obtained from a TM Landsat-5 classification. The adaptative filters showed the best results for speckle suppression and border preservation, especially the Kuan, Gamma MAP, Lee, Frost and Enhanced Frost filters. Small similarity and area thresholds (5 and 10, respectively) were used for the image segmentation due to the reduced dimensions and the narrow and elongated forms of the reefs. The classification threshold of 99% had a better user's accuracy, but a lower producer's accuracy because it is a more restrictive threshold; therefore, it may be possible that it had a greater omission on reef classification. The results indicate that SAR images have a good potential for the detection of emergent coral reefs.


2004 ◽  
Vol 120 ◽  
pp. 697-704
Author(s):  
L. Depradeux ◽  
J.-F. Jullien

In this study, a parallel experimental and numerical simulation of phenomena that take place in the Heat Affected Zone during TIG welding on 316L stainless steel is presented. The aim of this study is to predict by numerical simulation residual stresses and distortions generated by the welding process. For the experiment, a very simple geometry with reduced dimensions is considered: the specimens are disks, made of 316L. The discs are heated in the central zone in order to reproduce thermo-mechanical cycles that take place in the HAZ during a TIG welding process. During and after thermal cycle, a large quantity of measurement is provided, and allows to compare the results of different numerical models used in the simulations. The comparative thermal and mechanical analysis allows to assess the general ability of the numerical models to describe the structural behavior. The importance of the heat input rate and material characteristics is also investigated.


2012 ◽  
Vol 182-183 ◽  
pp. 541-545 ◽  
Author(s):  
Qi Ju Zhu ◽  
Gong Min Yan ◽  
Peng Xiang Yang ◽  
Yong Yuan Qin

A new rapid computation method for Kalman filtering is proposed. In this method, the prediction of state covariance matrix is expanded directly rather than computing by a looping program. Sequential filtering for measurement update is also applied. Furthermore, the subsidiary elements in system matrix are set to zero and a reduced-dimensions sub-optimal Kalman filter is presented. The proposed method greatly decreases computational burden and it is only 6.59% of the classic method. In the end, a vehicular test is carried out to prove the feasibility of the filtering.


MRS Bulletin ◽  
1999 ◽  
Vol 24 (8) ◽  
pp. 13-19 ◽  
Author(s):  
S.M. Prokes ◽  
Kang L. Wang

In recent years, tremendous interest has been generated in the fabrication and characterization of nanoscale structures such as quantum dots and wires. For example, there is interest in the electronic, magnetic, mechanical, and chemical properties of materials with reduced dimensions. In the case of nanoscale semiconductors, quantum effects are expected to play an increasingly prominent role in the physics of nanostructures, and a new class of electronic and optoelectronic devices may be possible. In addition to new and interesting physics, the formation and characterization of nanoscale magnetic structures could result in higher-density storage capacity in hard disks and optical-recording media. Likewise, phonon confinement leads to a drastic reduction of thermal conductivity and can be used to improve the performance of thermoelectric devices.In 1980, H. Sakaki predicted theoretically that quantum wires may have applications in high-performance transport devices, due to their sawtoothlike density of states (E1/2), where E is the electron energy. Since then, most quantum wires have been made by fabricating a gratinglike gate on top of a two-dimensional (2D) electron gas contained in a semiconductor heterojunction or in metal-oxide-semiconductor structures. By applying a negative gate voltage to the system, its structure can be changed from a 2D to a one-dimensional (1D) regime, where electron confinement is achieved by an electrostatic confining potential. It was not until recently that “physical” semiconductor quantum wires with the demonstrated 1D confinement by physical boundaries began to be fabricated.


Author(s):  
M. Wada ◽  
S. Mimura ◽  
H. Nihira ◽  
H. Iizuka

Sign in / Sign up

Export Citation Format

Share Document