14532 research outputs found
Sort by
Privacy-preserving with Flexible Autoencoder for Video Coding for Machines
No full textThe dataset for Video Coding for Machines (VCM) contains sensitive information that requires privacy preservation to address vulnerabilities. Achieving a balance to protect this sensitive data while maintaining VCM performance is crucial. We introduce an autoencoder integrated with a deep learning network that utilizes the ResNet architecture. This design blurs private details while preserving the contours, offering a high-dimensional representation that upholds privacy and VCM performance. The division position between the encoder and decoder is critical, influencing the equilibrium between compression efficacy and machine task performance. We craft a flexible, position-adjustable setting for the autoencoder to optimize this, facilitating a harmonious trade-off between bitrate and mAP across various deep-learning networks. This adaptation demonstrates superior performance relative to existing models. With FasterRCNN, our methods achieve 62.3 of mAP and 5681.29 of bitrate, and their versatility is further validated using YoloV5 and SSD.2024 IEEE International Symposium on Circuits and Systems (ISCAS)
Date of Conference: 19-22 May 202
Studies on Screening and Evaluation of Novel Compounds Focusing on Plants Stress Response Pathways
Get PDF横浜国立大学博士（理学
Enhancing Single-Electron Reservoir Computing Performance with Delay Function and Multiple-layer Reservoir Circuits
Get PDFWe have previously designed single-electron reservoir computing (SERC) circuit and have confirmed its competence for waveform prediction. In this study, we observed that the virtual increase in connections from the reservoir to the output by implementing delay functions led to an improvement in the performance for waveform prediction. Furthermore, the SERC circuit with multiple-layer reservoir possessed the capability to predict complex waveforms, which our previous circuit could not deal with.IEEE Silicon Nanoelectronics Workshop 2024
June 15-16, 2024
Hilton Hawaiian Village, Honolulu, HI, US
A Quasi-3-D Finite Element Modeling of An Axial Flux Magnetic Resonant Motor
Get PDFDue to inherent the 3-D magnetic structure of axial flux machines (AFM), 3-D finite element analysis (FEA) is the most commonly utilized approach for simulation in the design and analysis. However, these simulations are exceedingly time-demanding, particularly during the early phases of the design. This challenge has not been properly addressed in the design of an axial flux magnetic resonant motor (AFMRM). This in turn has limited the level of optimization that has been achieved in the design of AFMRM motor. In this paper, we present a method to quasi-3-D conduct FEA modeling analysis of an AFMRM motor. It is achieved by slicing the 3-D geometries of the AFMRM machine with cylindrical planes of different radii to form equivalent 2-D linear machine (2DLM) models. The magnetic problem is then solved in each slice with a 2-D finite element method (FEM). As a result, the generated mesh required for computation is faster and many designs analysis can therefore be implemented for optimization.2024 IEEE International Magnetic Conference
Rio de Janeiro, Brazil
05-10 May 202
A study on the relationship between the health of the elderly and the built environment in a hilly suburban residential area:Focusing on frailty of elderly
Get PDF横浜国立大学博士（学術）この学位論文の全文は、中央図書館で平日17時までに申請することで閲覧が可能です
