中国科学院机构知识库网格系统: 检索

浏览/检索结果: 共7条，第1-7条

帮助

条数/页：排序方式：
	Towards Better Generalization of Deep Neural Networks via Non-Typicality Sampling Scheme 期刊论文 OAI收割 IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 页码: 11 作者: Peng, Xinyu; Wang, Fei-Yue; Li, Li \| 收藏 \| 浏览/下载：34/0 \| 提交时间：2022/06/06 Training Estimation Deep learning Standards Optimization Noise measurement Convergence Deep learning generalization performance nontypicality sampling scheme stochastic gradient descent (SGD)
	An index tracking model with stratified sampling and optimal allocation 期刊论文 OAI收割 APPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY, 2018, 卷号: 34, 期号: 2, 页码: 144-157 作者: Wang, Meihua; Xu, Fengmin; Dai, Yu-Hong \| 收藏 \| 浏览/下载：33/0 \| 提交时间：2018/07/30 index tracking out-of-sample performance stratified sampling stratified hybrid genetic algorithm s-rar crossover
	Testing inequality constraints in a linear regression model with spherically symmetric disturbances 期刊论文 OAI收割 JOURNAL OF SYSTEMS SCIENCE & COMPLEXITY, 2014, 卷号: 27, 期号: 6, 页码: 1204-1212 作者: Zhu Rong; Zhou Sherry Z F \| 收藏 \| 浏览/下载：20/0 \| 提交时间：2021/01/14 SQUARED ERROR LOSS SAMPLING PERFORMANCE RESTRICTED ESTIMATOR HYPOTHESIS PRETEST COEFFICIENTS PARAMETER VARIANCE PRICES Inequality constraint spherically symmetric distribution Wald test
	Calibration and field performance of triolein embedded acetate membranes for passive sampling persistent organic pollutants in water 期刊论文 OAI收割 Environmental Sciences & Ecology, 2012, 卷号: 164, 期号: 1, 页码: 158-163 Tang, Jianfeng; Chen, Shan; Xu, Yiping; Zhong, Wenjue; Ma, Mei; Wang, Zijian 收藏 \| 浏览/下载：23/0 \| 提交时间：2014/11/28 Calibration Triolein embedded cellulose acetate membranes TECAM Performance reference compounds PRCs Passive sampling
	Design of driving circuit for binocular CCD image system (EI CONFERENCE) 会议论文 OAI收割 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Detector, Imager, Display, and Energy Conversion Technology, April 26, 2010 - April 29, 2010, Dalian, China 作者: Zhang X.; Zhang X.; Zhang X. 收藏 \| 浏览/下载：45/0 \| 提交时间：2013/03/25 The paper designs a driving circuit of high sensitive wide dynamic and high signal-to-noise ratio for binocular CCD imaging system which adopts a Dalsa-made high resolution full-frame 33-mega pixels area CCD FTF5066M. Inner structure and driving timing of the FTF5066M sensor are presented. Field Programmable Gate Array (FPGA) is used as the main device to accomplish the timing design of the circuits and power driver control of the two sensors. By using the Correlated Double Sampling (CDS) technique the video noise is reduced and the SNR of the system is increased. A 12- bit A/D converter is used to improve the image quality. The output rate of the imaging system designed with integrated chip can reach to 1.3 frames per second through bi-channel. For its good performance low power consumption and small volume the driving system can be applied to aeronautics and astronautics field. With a further improvement a maximum data output rate of 2.7 frames per second can be reached through all the eight channels of the two CCDs. 2010 Copyright SPIE - The International Society for Optical Engineering.
	A new driving method for LCoS with frame buffer pixels (EI CONFERENCE) 会议论文 OAI收割 ICO20: Display Devices and Systems, August 21, 2005 - August 26, 2005, Changchun, China Song Y.; Ling Z. 收藏 \| 浏览/下载：27/0 \| 提交时间：2013/03/25 A new driving method for LCoS microdisplays with frame buffer pixels was developed here. The power dissipation of the LCoS microdisplays with frame buffer pixels is higher than that of the LCoS with DRAM-like pixels due to the twice samplings in LCoS with frame buffer pixels. In this paper an adiabatic charging method was used to the second sampling of the frame buffer pixels in order to reduce the power dissipated in the transistors. The power dissipation of the second sampling was calculated when the power sources of the step the ramp and the stair-step are used respectively. The conventional design adopted the step and contributed to high power and so result in more heat to deteriorate the device performance. The power dissipated in the transistors is almost zero if the ideal ramp source is used. The ramp can be the stair-step whose steps are infinity. The stair-step substitutes for the ramp due to easily generation and higher energy efficiency than the step. It can decrease the power dissipation of the LCoS panel and contributes to the heat reduction caused by power dissipation which can increase the microdisplay devices reliability. This method was developed based on the frame buffer pixel circuits which we proposed previously and can be applied to the others.
	Geometrical modulation transfer function of different active pixel of CMOS APS (EI CONFERENCE) 会议论文 OAI收割 2nd International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment, November 2, 2005 - November 5, 2005, Zian, China 作者: Li J.; Liu J.; Liu J.; Liu J.; Li J. 收藏 \| 浏览/下载：30/0 \| 提交时间：2013/03/25 The geometrical Modulation Transfer Function (MTF) of CMOS APS (active pixel sensor) is analyzed in this paper. Advanced APS have been designed and fabricated where different pixel shapes such as square rectangle and L shape were placed because the amplifier circuit and other function circuits inter pixel of APS take up some pixel area. MTF is an important figure of merit in focal plane array imaging sensors. Research on analyzing the MTF for the proper pixel shape is currently in progress for a centroidal configuration of a target position. MTF will give us a more complete understanding of the tradeoffs opposed by the different pixel designs and by the signal processing conditions. Based on image sensor sampling and reconstructing model the MTF expression of any active pixel shape has been deduced in this paper. According to actual pixel shape three different active area pixels were analyzed they were square rectangle and L shape their Fill Factor (FF) is 30% 44% and 55% respectively. Results of simulation experiments indicate that different pixel geometrical characteristics contribute significantly to the figures of their MTF. Different geometrical shape of active sensitive area of pixel and different station in pixel would influence MTF figures. The analysis results are important in designing better APS pixel and more important in analyzing imaging system performance of APS subpixel precision system.