University of Electro-Communications

About: University of Electro-Communications is a education organization based out in Tokyo, Japan. It is known for research contribution in the topics: Laser & Robot. The organization has 8041 authors who have published 16950 publications receiving 235832 citations. The organization is also known as: UEC & Denki-Tsūshin Daigaku.

Dynamic recrystallization under warm deformation of a 304 type austenitic stainless steel

Abstract: Warm (and hot) deformation of a 304 type austenitic stainless steel was studied in connection with microstructural developments in compression at temperatures of 873–1223 K (0.5–0.7 Tm) under strain rates of 10−4–10−1s−1. The two deformation domains can be categorized due to their different mechanical and microstructural behaviors. In the region of flow stresses lower than around 400 MPa, the deformation behaviors are typical for hot working accompanied with dynamic recrystallization (DRX). New grains are evolved mainly by dynamic bulging mechanism, which can be accelerated by the development of serrated grain boundaries and strain induced dislocation subboundaries. The relationship between dynamic grain sizes ranged from 2 to 7 μm and peak flow stress can be expressed by a power law function with a grain size exponent of −0.72. In contrast, in the region of flow stresses higher than 400 MPa, the deformation behaviors hardly depend on strain rate and temperature and so can be in the region of athermal deformation. The stress–strain curves under such warm deformation are similar to those affected only by dynamic recovery. The microstructures evolved at high strains are mainly characterized by the dense dislocation walls evolved in pancaked original grains, while grain boundary serration also takes place even at such warm deformation. Mechanisms of this microstructural evolution are discussed in combination with analysis of deformation mechanisms operating under warm deformation.

A food image recognition system with Multiple Kernel Learning

Abstract: Since health care on foods is drawing people's attention recently, a system that can record everyday meals easily is being awaited. In this paper, we propose an automatic food image recognition system for recording people's eating habits. In the proposed system, we use the Multiple Kernel Learning (MKL) method to integrate several kinds of image features such as color, texture and SIFT adaptively. MKL enables to estimate optimal weights to combine image features for each category. In addition, we implemented a prototype system to recognize food images taken by cellular-phone cameras. In the experiment, we have achieved the 61.34% classification rate for 50 kinds of foods. To the best of our knowledge, this is the first report of a food image classification system which can be applied for practical use.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry using an inorganic particle matrix for small molecule analysis

Abstract: Fine metal or metal oxide powder as an alternative to conventional organic matrices in matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) has been utilized successfully for lower molecular mass analytes, poly(ethylene glycol) 200 (PEG 200) and methyl stearate. Eleven kinds of particle, Al, Mn, Mo, Si, Sn, SnO2, TiO2, W, WO3, Zn and ZnO, were evaluated. The analyte was mixed with a metal or metal oxide powder (inorganic matrix) with particle diameter of tens of micrometers and liquid dispersant, followed by application to the sample target. Using a commercial MALDI-TOFMS instrument equipped with an internal 337 nm pulsed nitrogen laser, the analytes, PEG 200 and methyl stearate, were ionized as the alkali metal ion adducted molecules [M+Na]+ or [M+K]+ when the inorganic matrices Mn, Mo, Si, Sn, TiO2, W, WO3, Zn or ZnO were used. In the case of an Al matrix, PEG 200 was ionized as [M+K]+, whereas methyl stearate was ionized as [M+H]+ and [M+Al]+. These particles have potential as the matrix for MALDI. During our examination, however, only SnO2 particles did not ionize either PEG 200 or methyl stearate. Based on our protocol, when TiO2 powder was suspended with liquid paraffin, PEG 200 and methyl stearate gave their MALDI-TOF mass spectra with the lowest background noise and highest intensity. TiO2 powder seemed to be a broad potential matrix for low molecular mass polar or non-polar analytes. The results suggested that bulk particles caused rapid heating/vaporization processes and ionized analyte molecules under irradiation with a pulsed UV laser. The present method can be readily applied to obtain the low background noise MALDI-TOF mass spectra of small-sized compounds.

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

Abstract: Federated learning (FL) is a distributed machine learning approach that can achieve the purpose of collaborative learning from a large amount of data that belong to different parties without sharing the raw data among the data owners. FL can sufficiently utilize the computing capabilities of multiple learning agents to improve the learning efficiency while providing a better privacy solution for the data owners. FL attracts tremendous interests from a large number of industries due to growing privacy concerns. Future vehicular Internet of Things (IoT) systems, such as cooperative autonomous driving and intelligent transport systems (ITS), feature a large number of devices and privacy-sensitive data where the communication, computing, and storage resources must be efficiently utilized. FL could be a promising approach to solve these existing challenges. In this paper, we first conduct a brief survey of existing studies on FL and its use in wireless IoT. Then, we discuss the significance and technical challenges of applying FL in vehicular IoT, and point out future research directions.

Minimal electronic models for superconducting BiS 2 layers

Abstract: We construct minimal electronic models for a newly discovered superconductor LaO${}_{1\ensuremath{-}x}$F${}_{x}$BiS${}_{2}$ (${T}_{c}=10.6$ K) possessing BiS${}_{2}$ layers based on a first-principles band calculation. First, we obtain a model consisting of two Bi $6p$ and two S $3p$ orbitals, which give nearly electron-hole symmetric bands. Further focusing on the bands that intersect the Fermi level, we obtain a model with two $p$ orbitals. The two bands (per BiS${}_{2}$ layer) have a quasi-one-dimensional character with a double minimum dispersion, which gives good nesting of the Fermi surface. At around $x\ensuremath{\sim}0.5$ the topology of the Fermi surface changes, so that the density of states at the Fermi level becomes large. Possible pairing states are discussed.