Development of an image evaluation method for skin color distribution in facial images and its application: Aging effects and seasonal changes of facial color distribution

In this article, we present an evaluation method for the skin color distribution in the face area. Unlike previous methods that extract a specific area, our method subdivides the entire face into small regions and analyzes detailed, per-frame textures. Our evaluation method for skin color distribution is based on facial feature points and includes segmentation that takes into account the facial skeleton and muscle orientation. The use of facial feature points enables a comparison at relatively equal positions on the face without depending on the shape or size of the individual's face. Our evaluation method is versatile, and as an application, we clarified age-specific features and seasonal variations of facial color distribution. As a result of applying this evaluation method to the facial images of women aged 20 to 78 years, we confirmed that the lightness of the face decreased as age increased. In particular, the decrease in lightness was remarkable in the region along the cheekbone, from the temple to the center of the cheek. Furthermore, we analyzed the seasonal changes of melanin distribution in the face area. This showed that the melanin index increased particularly in the cheekbone area in the summer when the influence of ultraviolet rays became large. Our novel methodology and the data presented in this article will be useful in various fields, such as dermatology, cosmetics, and computer vision.     

Development of Sensor and Control Systems of a Wearable Robot to Walk on a Step

The goal of our study is to develop a system for walking on a step using a wearable robot. Our system consists of (1) sensing of a step from the movement of the walker, (2) detection of the foot placement state related to the step, and (3) generation of gait patterns of climbing and stepping down for the step. In the generation of gait patterns for the step, toe trajectories are generated according to the height of the step to avoid collision of the swinging leg with the step. The hip trajectory is generated by an optimization technique that minimizes the sum of the joint angular jerk of the robot subject to constraints on the hip position and the velocity at toe liftoff. Each joint angle trajectory is calculated from the generated trajectories by means of inverse kinematic equations. We investigated the feasibility of the proposed sensor and control systems for two steps with different heights.     

Reversible Reaction with CO2 and Control of CO2 Absorption/Desorption Properties of Ba2(Fe1−xInx)2O5 Solid Solutions

Ba₂(Fe_1?xIn_x)₂O₅ was prepared by a solid‐state reaction under a N₂ flow. It was revealed that the solid solutions had a cubic perovskite structure with disordered oxygen vacancies at room temperature. Thermogravimetry and X‐ray diffraction measurements revealed that Ba₂(Fe_1?xIn_x)₂O₅ can reversibly react with CO_2. It was found that the equilibrium temperature of the reaction could be controlled by preparing solid solution.     

Multi-scale parallel finite element analyses of LDH sheet formability tests based on crystallographic homogenization method

A multi-scale parallel finite element (FE) procedure based on the crystallographic homogenization method was applied to the LDH sheet formability test analysis. For the multi-scale structure, two scales are considered. One is a microscopic polycrystal structure and the other is a macroscopic elastic plastic continuum. The analysis code can predict the formability of sheet metal in macro-scale, simultaneously the crystal texture and hardening evolutions in the micro-scale (Nakamachi E et al. Int J Plasticity 2007;23:450-8). Since huge computation time is required for the nonlinear dynamic multi-scale FE analysis, parallel computing technique based on domain partitioning of FE model for macro-continuum is introduced into the multi-scale code using the message passing interface (MPI) library and PC cluster (Kuramae H et al. In: Proceedings of the eighth international conference on computational plasticity, Part 1, 2005. p. 622-5). The explicit time stepping solution scheme in the nonlinear multi-scale FE dynamic problem is well-suited for parallel computing on distributed memory environment such as PC cluster because solving simultaneous equation is not required. We measured crystal morphologies of four automotive sheet metals, aluminum alloy sheet metals A6022-T43 and A5182-O, an asymmetrically rolled aluminum alloy sheet metal A6022-ASR, and mild steel HC220YD, by using the scanning electron microscope (SEM) with electron back scattered diffraction (EBSD) analyses, and defined a three-dimensional representative volume element (RVE) of micro polycrystal structure, which satisfy the periodicity condition of crystal orientation distribution. We evaluate not only macroscopic formability of the automotive sheet metals by the multi-scale LDH test analysis, but also microcrystalline texture evolution during plastic deformation. Furthermore, a relationship between the macroscopic formability and the microcrystal texture evolution was discussed through looking at multi-scale FE results. It is concluded that the mild steel HC220YD was the highest formability than the aluminum alloy sheet metals because of remaining and generating the γ-fiber texture, such as {1 1 1}〈1 1 0〉-{1 1 1}〈1 1 2〉 orientations, during plastic deformation.     

Aluminium content of foods originating from aluminium-containing food additives

Aluminium (Al) levels of 90 food samples were investigated. Nineteen samples contained Al levels exceeding the tolerable weekly intake (TWI) for young children [body weight (bw): 16 kg] when consuming two servings/week. These samples were purchased multiple times at specific intervals and were evaluated for Al levels. Al was detected in 27 of the 90 samples at levels ranging from 0.01 (limit of quantitation) to 1.06 mg/g. Of these, the Al intake levels in two samples (cookie and scone mix, 1.3 and 2 mg/kg bw/week, respectively) exceeded the TWI as established by European Food Safety Authority, although the level in the scone mix was equivalent to the provisional TWI (PTWI) as established by Joint Food and Agriculture Organization of the United Nations/World Health Organization Expert Committee on Food Additives. The Al levels markedly decreased in 14 of the 19 samples with initially high Al levels. These results indicated reductions in the Al levels to below the PTWI limits in all but two previously identified food samples.     

Attentional modulation of the mere exposure effect.

The mere exposure effect refers to the phenomenon where previous exposures to stimuli increase participants’ subsequent affective preference for those stimuli. This study explored the effect of selective attention on the mere exposure effect. The experiments manipulated the to-be-attended drawings in the exposure period (either red or green polygons in Experiments 1 and 2; both red and green polygons in Experiments 3 and 4) and black to-be-evaluated drawings in the affective judgment period (morphologically identical to the red or green polygons in Experiments 1 and 4; morphologically identical to the composite drawings in Experiments 2 and 3). The results showed a significant mere exposure effect only for the target shapes involved in attentional selection, even when the participants could recognize the nontarget shapes. This indicates that selective attention modulated the mere exposure effect. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Thickness estimation of interface films formed on Li1−xCoO2 electrodes by hard X-ray photoelectron spectroscopy

Solid electrolyte interface (SEI) films formed on Li_1−xCoO₂ electrodes were observed with hard X-ray photoelectron spectroscopy (HX-PES). This paper particularly focuses on film thickness estimation using HX-PES with theoretical calculation. The validity of the calculation was proven by experiments using model SEI films. The native film formed on a LiCoO₂ composite electrode was estimated to be LiF with its thickness of 5 nm. Formation of Co (II) species on top of LiCoO₂ was also indicated. Storage of the electrode at 60 °C brought about considerable film growth (30-40 nm) with carbonate compounds formation. SEI film changes during charging of the LiCoO₂ electrode were also examined. The main component in the film was deduced to be LiF or a kind of fluorite, with its thickness decreased during charging. The SEI formation mechanisms are also elucidated.