Search

The thermoelectric hydrogen (H2) gas sensor using PCB (Printed Circuit Board) technique was developed based on the exothermic absorption of H2 by a palladium (Pd) film coated on a thermocouple. A cascade connection of two thermocouples composed of copper (Cu) and constantan (55% nickel and 45% Cu) was used to detect the exothermic absorption of H2 by Pd. The differential thermoelectric voltage output between the two thermocouples (with and without the Pd film) increased linearly with the H2 gas concentration in a 2.0–50 vol% ambient atmosphere. Standard deviations (SD) for 8 measurement cycles are typically 1.1% at 4 vol% H2, 0.81% at 6 vol% H2, and 1.9% at 8 vol% H2, respectively. The differential thermoelectric voltage output can be detected from 2.0 vol% H2 atmosphere. The ambient temperature fluctuations on the measurement was also effectively reduced using the cascade connection of two thermocouples. Calibration line of H2 concentration calculated by the least square method is linear and standard error (SE), 0.44 vol%, is smaller than measured value.

An 80-year-old man with a history of cutaneous squamous cell carcinoma (SCC) was referred to our department for a solitary lung nodule. The nodule was surgically resected and diagnosed as SCC. Because the lung lesion and a previous skin lesion showed similar histological findings, the origin of the lung tumor was uncertain. Next-generation sequencing using a targeted driver oncogene panel was applied for the further examination. The lung lesion was diagnosed as primary lung SCC, as the two tumors possessed distinct somatic mutations in TP53. Recent advances in clinical sequencing have enabled us to obtain an accurate diagnosis in pathologically challenging cases.

Optical scattering spectra and angular distribution of scattering intensities were evaluated using concentrated orange juice. The concentrated orange juice was opaque due to the scattering particles included. In these specimens, the scattering intensities and angular distribution of the scattering intensities varies with the refractive index ratio of the scattering particles against the liquid surrounding the particles. The sugar concentrations of orange juice, which are linearly correlated with the refractive index of the specimens, can therefore be measured by the scattering intensities and their angular distributions. The scattering intensities measured using a halogen lamp, a white LED, a yellow LED, a red LED and near-infrared LEDs (λ = 850 nm and 940 nm) vary linearly with the sugar concentrations of the specimens. The gradient and correlation coefficient of the calibration lines fitted are found to vary with the scattering angles, θ. A non-destructive evaluation of the sugar concentrations of orange juice can be measured sensitively at θ = 170 or 160 °s with any incident lights which are scattered from the specimens.

The ultraviolet C rays (UVC, wavelength λ = 100–280 nm) light generated by a Hg lamp ( λ = 254 nm) and UVC light-emitting diodes (LEDs, λ = 265 and 275 nm) was detected using a fluorescence-enhanced silicon photodiode (FE-PD). Ce-doped yttrium aluminum garnet (YAG:Ce), YAG:Pr, YAG:Eu, YAG:Tb, YAG:Cr, Al2O3:Ti, Al2O3:Cr, MgAl2O4:Ti, MgAl2O4:Cr, MgAl2O4:Mn, and commercial fluorescent acrylic resins were tested as phosphor sources to enhance the output signal intensity of the FE-PD irradiated with UVC light. The resulting output signal intensity increased linearly with the UVC light strength, which was adjusted by raising the input current of the UVC LEDs from 0 to 40 mA. The sensitivity of the fabricated UVC detectors, assessed based on the calibration curve slope, varied depending on the phosphor materials. The phosphors effectively enhanced the output signal intensity of the FE-PD, which was up to six times greater than that of the visible and near infrared Si-PD without phosphors; the stronger output signal intensity was achieved using YAG:Tb, YAG:Cr, and a red fluorescent acrylic resin. The visible light emitted by phosphors under UVC irradiation is useful for detecting UVC light by the eye when using FE-PD.

The next step for the generative adversarial networks (GAN) is to learn representations that allow us to control only a certain factor in the image explicitly. Since such a representation of the factor is independent of other factors, the controllability obtained from these representations leads to interpretability by identifying the variation of the synthesized image and the transferability for downstream tasks by inference. However, since it is difficult to identify and strictly define latent factors, the annotation is laborious. Moreover, learning such representations by a GAN is challenging due to the complex generation process. Therefore, we resolve this limitation using a novel generative model that can disentangle latent space into the appearance, the x-axis, and the y-axis of the object, and reassemble these components in an unsupervised manner. Specifically, based on the concept of packing the appearance and location in each position of the feature map, we introduce a novel structural constraint technique that prevents these representations from interacting with each other. The proposed structural constraint promotes the disentanglement of these factors. In experiments, we found that the proposed method is simple but effective for controllability and allows us to control the appearance and location via latent space without supervision, as compared with the conditional GAN.

Rendering an any-viewpoint image is extremely difficult for Generative Adversarial Networks. This is because conventional GANs do not understand 3D information under-lying a given viewpoint image such as an object shape and relationship between viewpoint and objects in 3D space. In this paper, we present how to perform a Viewpoint-Agnostic Image Rendering (VAIR), equipping a conditional GAN with a mechanism to reconstruct 3D information of the input view. VAIR realizes any-viewpoint image generation by manipulating a viewpoint in 3D space where the reconstructed instance shape is arranged. In addition, we convert the reconstructed 3D shape into a 2D representation for image-based conditional GAN, while preserving detail 3D information. The representation consists of a depth image and 2D semantic keypoint images, which are obtained by rendering the shape from a viewpoint. In the experiment, we evaluate using a CUB-200-2011 dataset, which contains few-samples biased a viewpoint such that covers only part of the target appearance. As a result, our VAIR clearly renders an any-viewpoint image.

In this paper, we propose convolutional neural networks for semantic segmentation on road markings in the situation where sequential segmentation ground truth masks are available. The proposed model aggregates the temporal information and the context information from the multiple frames. Moreover, we employ CGNet as the backbone network to reduce trainable parameters and computation speed. In the experiment, we evaluate the model using the Gifu-city Road Marking Segmentation Dataset, which includes road markings of open roads in Gifu city. As a result, the segmentation performance such as a white center line and white dash line is an improvement.

Mn doped spinel (MgAl 2 O 4 ) crystals have been grown by floating zone (FZ) techniques with various conditions of O 2 concentrations in the growth atmosphere from 0 to 75 vol%, with a starting composition of molar ratio x= MgO/(MgO+Al 2 O 3 ) from 0.3 to 1.0 and/or Mn concentrations from 1.0 to 6.0 at%. Optical absorption spectra and photoluminescence spectra were evaluated using crystals grown under various growth conditions. The color of the crystals is found to vary with their composition and the O 2 concentration in the atmosphere. Crystals grown under a 100 vol% Ar atmosphere, were a pale green color and emitted a strong green luminescence at λ =520 nm. Colors of stoichiometric crystals, x =1.0, and Mg-poor crystals grown under an oxidizing atmosphere are yellow and red, respectively. Both optical absorption spectra and photoluminescence spectra of Mg-poor, x 2 –Ar) atmosphere are found to vary greatly from those of the stoichiometric crystals.

Designing the optimal architecture of neural networks is an important issue. However, since this is difficult even for experienced experts, automatic optimization of the network architecture is required. In this study, we regard this issue as a combinatorial optimization problem, and utilize genetic algorithm to optimize the network architecture. Because training the networks, which are represented by individuals in GA, takes a long time, a novel method to reduce the training time by inheriting the weights of the trained network is proposed. From experimental results, our proposed method achieved the time reduction and higher accuracy than a conventional method.

We propose a defect detection method of solders on a printed circuit board using X-ray CT inspection system and Adversarial Autoencoder (AAE)[1] . We obtain sliced images of the solder using X-ray CT and extract their features that follow the standard normal distribution by using AAE. Then, the solder defects are detected by Hotelling's T square[2]. As a result of experiments, we show that we can classify normal and anomalous data samples completely on the condition of training with large normal samples and small anomalous samples.

X-ray excited visible light luminescence were evaluated from Cr doped Y3Al5O12 (YAG) crystals using an X-ray powder diffractometer adapting a fiber spectrometer, a digital camera, and a LED illumination. Red colored luminescence peaking at λ = 678, 690, 708, and 726 nm were observed from Cr doped YAG under irradiation by both CuKα X-ray and a violet LED (λ = 405 nm). Peak wavelengths of X-ray excited luminescence are the same as those of photoluminescence (PL). The peak intensities of X-ray excited luminescence, which increases linearly with X-ray strength, increases up to Cr concentration at around 0.5 mol%, then decreases gradually with Cr concentration due to a concentration quenching. It was found that the concentration quenching of X-ray excited luminescence occurs at a lower concentration at around 0.5 mol% than that of PL at around 3.5 mol%.

A fluorescent temperature sensor using temperature dependence of fluorescent inorganic materials is studied for low temperature measurement. Ruby (chromium ions doped sapphire) is used as a sensor head of the fluorescence thermometer because of long photoluminescence (PL) lifetime and visible emission. In this research, ruby crystal is found to be useful as the sensor head for fluorescence thermometer. In addition, the ruby crystal is potentially useful material for the sensor head crystals to measure low temperature below 0 °C.

Concentrations of vitamin B2 in milk were evaluated based on photoluminescence (PL) peak intensities at around λ = 545 nm. PL was measured in a back-scattering configuration using a 3.2 mm diameter optical fibre bundle irradiated by ultra-violet LEDs, violet LEDs or a blue LED as the excitation light. The tip of the optical fibre bundle was dipped into commercially available whole milk, low-fat milk and non-fat milk. PL peaking at 545 nm was observed under LED irradiation. Vitamin B2 concentrations were measured using a standard addition method and a calibration curve method. Deviation between the vitamin B2 concentration measured by PL using a standard addition method and those by HPLC technique varied with wavelength of excitation light and was smallest with violet LED (405 nm) excitation. The standard deviation of results measured by PL using a calibration curve method was better than those using a standard addition method.

Optical back-scattering from milk with various fat and protein concentrations were evaluated in a back-scattering configuration using white, violet, blue, green, red or near infrared (NIR, λ = 850 and 940 nm) LEDs as the light source. The tip of the optical fibre bundle was dipped into the milk with fat and protein concentrations from 8.2 to 39 g L−1 and 32 to 37.7 g L−1, respectively. The back-scattering intensities varied linearly with the fat and protein concentrations of the milk. The slope (G) of calibration lines varied with the wavelength of the incident light. The G of the fat concentrations became zero at wavelength around λ = 520 nm and that of the protein concentrations was much smaller than that of the fat concentrations at NIR wavelengths. The fat and protein concentrations can therefore be measured separately based on the scattering intensities by selecting the wavelength for measurement.

As corals in tropical regions are threatened by increasing water temperatures, poleward range expansion of reef-building corals has been observed, and temperate regions are expected to serve as refugia in the face of climate change. To elucidate the important indicators of the sustainability of coral populations, we examined the genetic diversity and connectivity of the common reef-building coral Acropora hyacinthus along the Kuroshio Current, including recently expanded (

Luminescence properties of Mn doped MgAl

In this paper, we discuss a method to detect defects in industrial products by using Denoising AutoEncoder Generative Adversarial Networks. In previous methods, a defective area is detected by restoring a defective product image which added an artificial defect to a non-defective product image by Denoising AutoEncoder (DAE). Therefore, a defective area is detected by subtracted image of them. We discuss whether further accuracy improvement is possible by introducing a framework of adversarial learning to DAE in order to restore a defective image to a non-defective image clearer.

A novel thermoelectric type hydrogen gas sensor based on exothermic reaction of hydrogen storage alloy is proposed and fabricated. First, an exothermic reaction of a hydrogen storage alloy at room temperature has been studied for a hydrogen gas sensor application. Next, a copper (Cu) constantan thermocouple painted a palladium (Pd) past on surface was fabricated as a hydrogen gas detector. The response characteristic of the fabricated hydrogen gas detector was evaluated under hydrogen gas flow. The exothermic reaction of Pd and hydrogen gas was successfully observed using fabricated gas detector. It was clarified that the fabricated hydrogen gas detector using Cu constantan thermocouple with Pd is potentially useful for hydrogen gas sensor material.

Spinel (MgAl 2 O 4 ) crystals doped with 1.0% Mn have been grown by floating zone (FZ) technique with various Mg compositions, x = MgO/Al 2 O 3 , from 0.2 to 1.0. Compositional variations of photoluminescence are evaluated for a fluorescence thermometer application using crystals grown. Strong photoluminescence (PL) peak is observed at λ from 512 to 520 nm from the crystals grown from compositions, x , from 0.3 to 1.0. Peak wavelength of PL increases linearly from 512 to 520 nm with x . Weak PL peaking at λ = 750 nm is also observed from the specimens. Compositional variations of PL are considered to be due to the variation of crystal field surrounding the Mn 2+ ions. The variation of crystal field strength agrees with the compositional variation of lattice constant.

Eu 2 O 3 doped Y 3 Al 5 O 12 (YAG) crystals have been grown using a floating zone technique and evaluated thermal stability and annealing behavior of PL for a fluorescence thermo-sensor application. Color of the crystals grown varies from deep purple to colorless with O 2 concentration of the growth atmosphere and annealing in air. Photoluminescence (PL) peaking at λ = 590, 607, 624, 647 and 709 nm due to Eu 3+ ions are observed from the crystals under UV excitation. Anomalous temperature dependence of PL intensity, which is observed in as-grown crystals, is improved greatly by annealing through the heat cycle. From annealing behavior of optical absorption spectra, residual Eu 2+ ions are suggested to be responsible for the de-coloration and the improvement of anomalous temperature dependence of Eu doped YAG crystals.