Hosei University Repository
Not a member yet
22008 research outputs found
Sort by
PREPARATION OF [3-(ALKOXY)OLIGOSILANYL] LITHIUM AND ITS APPLICATION TO OLIGOSILANE SYNTHESIS
Get PDFIn this study I prepared [3-(alkoxy)trisilany]lithium 6 at -78℃ by direct reaction of chlorotrisilane 7 with lithium naphthalenide or by Sn-Li exchange reaction of stannyltrisilane 8 with n-buthyllithium. The reactions of 6 with chlorosilanes afforded tetrasilanes 12 and 13. The reactions of 6 with Cl-(SiMe_2)_n-Cl (n = 1, 2) afforded heptasilane 9 and octasilane 10. The UV absorption spectra of 9 and 10 were measured. The TD-DFT calculations revealed that the longest absorption wavelengths were attributed to the transition from HOMO to LUMO in 9 and 10
FDTD ANALYSIS OF AN INSB SPHERE ARRAY IN THE TERAHERTZ REGIME
Get PDFWe investigate a periodic array of an InSb sphere on a substrate at terahertz frequencies using the three-dimensional finite-difference time-domain method. We evaluate the transmission characteristics, paying attention to the resonance frequency. It is shown that the guided mode and surface plasmon resonances occur at the frequency dips
ESTABLISHMENT OF THE TECHNOLOGICAL BASIS FOR THE REALIZATION OF PHASED ARRAY WIRELESS CHARGING WHICH ENABLES MAGNETIC FIELD SUPPRESSION AT FOREIGN OBJECT LOCATION
Get PDFIn the wireless power transmission (WPT) to electric vehicles (EVs) in parking lots, there is a risk of abnormal heat generation due to absorption of the magnetic field in metallic foreign objects. Accordingly, currently available products are equipped with a function that automatically halts power transmission when a metallic foreign object is detected. However, if possible, continuing power transmission while suppressing the magnetic field absorption maybe another solution. Therefore, this paper proposes a novel function which enables wireless power transmission with a high efficiency while suppressing the magnetic field absorption of metallic foreign objects. In this study, it was assumed that a metallic foreign body is present at an arbitrary point in a two-dimensional plane, and the power transmission was conducted by the phased array WPT. An algorithm using the particle swarm optimization (PSO) to search for the optimal combination of the phase and amplitude of the coil input voltages together with coil arrangements in terms of both magnetic field suppression and transmission efficiency was proposed. The simulation was performed with the lower efficiency boundary set as 85% and the load power set as 11 kW, in reference to the SAE J2954 standard. As a result, it was confirmed that the magnetic field suppression effect increased in accordance with the increase in the number of transmission (Tx) coils, thus indicating the effectiveness of the proposed algorithm
Evolutionary dynamics optimization and FPGA based implementation of digital spike map
Get PDFIn this thesis, we consider optimization and FPGA based implementation of digital spike maps. First, the dynamics of spike-trains is visualized by a digital spike map. The map is defined on a set of points and is represented by a characteristic vector of integers. Second, we introduce a simple evolutionary algorithm for optimization of digital spike maps. We use autocorrelation function as a cost function. Third, in order to implement the digital spike map, we introduce a digital spiking neuron. Repeating integrate-and-fire behavior between a periodic base signal and constant threshold, the neuron can out put various periodic spike-trains. The digital spike maps are implemented in an FPGA board and typical spike-trains are confirmed experimentally
A Digital Direct Drive Speaker System with a Digital Feedback using Error Amplifier Circuit
Get PDFIn recent years, with the development of portable devices such as smartphones and tablets, there has been an increasing demand for audio systems with higher precision, smaller footprints, and lower power consumption. In this research, we have focused on improving the accuracy of our laboratory\u27s original audio system, the digital direct drive speaker system. Therefore, we proposed a feedback-type digital direct-drive speaker system using an error amplifier circuit at the circuit level. The accuracy requirement of the feedback ADC is relaxed, and the noise generated in the driver circuit is reduced. As a result, SNR of 107.3 dB and THD -97.5 dB were achieved
Dim-Light Robust Monocular SLAM : How Image Processing Technology Boosts the Robustness of Traditional Visual SLAM in Dim-Light Environment
Get PDFSLAM is the abbreviation of simultaneous localization and mapping. The mainstream methods of SLAM are Lidar SLAM and Visual SLAM. Compared with Lidar SLAM, Visual SLAM is cheaper, content distinguishable and easy to get. However, Lighting conditions are critical to the performance of visual SLAM system. Especially, challenge still remains for adopting visual SLAM in dim-light environment since it’s difficult to detect enough valid feature points. To address this issue, we propose DRMS (Dim-light Robust Monocular SLAM), a new method combining image preprocessing, which includes linear transformation and CLAHE, with the Monocular SLAM system. After applying the linear transformation and CLAHE, the brightness and contrast of the images would be significantly increased, and adequate feature points would be detected. Moreover, we use optical flow algorithm to track the features in order to reduce computation complexity. The performance of our method is validated both on public dataset and real-world experiment. The results show that our proposal is more reliable and of higher accuracy in dim-light conditions than other existing work
JAPANESE TEXT ANALYSIS CONSIDERING VISUAL AND SEMANTIC INFORMATION OF INTERPRETABLE CHARACTERS
Get PDFWhile Asian languages use characters with visual information such as radicals and shapes of kanji characters, japanese also uses characters without visual information such as hiragana and katakana. To take these features of japanese into account, a text analysis model that considers semantic information as well as visual information of characters has been proposed, and has achieved high performance in text analysis. However, while the visual information of the characters can be interpreted as radicals and shapes, the semantic information cannot be interpreted. In this paper, we propose text analysis model that considers highly interpretable visual and semantic information. The visual and semantic information of the character is separated and stored in each dimension of the low-dimensional representation of the character to enhance the interpretability. In addition, we focus on the fact that the acquisition of the meaning of characters in the semantic information depends on the training data. By combining the visual and semantic information, we expect to improve the robustness of the acquisition of the meaning of characters and to achieve high performance in text analysis. In sentiment analysis, we were able to interpret the semantic information of characters such as negative, positive, and user information. The combination of visual and semantic information improved the sentiment analysis task by about 1% and the recognize textual entailment task by about 0.3% compared to using only one of the two
Super-resolution for Brain MR Images from Significantly Small Amount of Training Data
Get PDFThis paper proposes two essential techniques to effectively train generative adversarial network-based super-resolution network for brain magnetic resonance images, even when only a small number of training samples are available. First, stochastic patch sampling is proposed, which increases training samples by sampling many small patches from the input image. However, sampling patches and combining them causes unpleasant artifacts around patch boundaries. The second proposed method, artifact-suppressing discriminator, suppresses the artifacts by taking two-channel input containing an original high-resolution image and a generated image. With the introduction of the proposed techniques, the network achieved to generate a natural-looking MR images from only ～ 40 training images, and improved the area-under-curve score on Alzheimer’s disease from 76.17% to 81.57%
ASYNCHRONOUS GATHERING ALGORITHMS FOR AUTONOMOUS MOBILE ROBOTS
Get PDFWe consider a Gathering problem for n autonomous mobile robots in an asynchronous scheduler (ASYNC). We use two models of the robots that have the capability of detecting whether there is more than one robot or not at its current point called local weak multiplicity detection and persistent memory called light. The result has been provided that Gathering can be solved with strong multiplicity detection if and only if n is odd. In the light model, it is known that Gathering can be solved by robots with 10 colors.This paper shows that Gathering can be solved by n = 3, 4 robots with local weak multiplicity detection. Additionally, we improve the result by reducing the number of colors 10 to 3. We also show that we can construct a simulation algorithm of any unfair SSYNC algorithm using k colors by ASYNC robots with 3k colors, where unfairness does not guarantee that every robot is activated infinitely often
AN ATTEMPT AT DOMAIN ADAPTATION IN AUTOMATIC PLANT DISEASE DIAGNOSIS SYSTEMS
Get PDFIn the automatic diagnosis system of plant diseases, there is a problem that the classification performance of the data obtained in the same field as the training data is high, but that of the data in the field not used for training is greatly reduced. The reason for this is that images taken in the same environment often have similar backgrounds, disease appearance, etc., and the same photographic equipment is often used, so the system overfitting on information specific to these fields. In recent years, the field of domain adaptation has been studied in order to maintain the performance on such different training data and data obtained from different distributions. In this paper, we implemented one unsupervised domain adaptation method and two domain standardization methods in a practical setting for the task of automatic plant disease diagnosis, and compared and verified the results. The results showed that the domain adaptation method improved the performance of the task by a small amount, but did not lead to a significant improvement. We also checked whether the current automatic plant disease diagnosis system has the information of the field where the images were taken, and investigated how much the current model depends on the information of the field. From the experiments, it was confirmed that the current disease classification model has a lot of field information and may be using the field information for disease classification