检索结果-内蒙古大学图书馆

48th International Conference on Computers and Industrial Engineering, CIE 2018

作者： Jia, Zhengxuan Xiao, Yingying Zhao, Zhilong Lin, Tingyu State Key Laboratory of Intelligent Manufacturing System Technology Beijing Institute of Electronic System Engineering Beijing China Beijing Complex Product Advanced Manufacturing Engineering Research Center Beijing Simulation Center Beijing China Science and Technology on Space System Simulation Laboratory Beijing Simulation Center Beijing China

With rapid development of machine learning, supervised learning and reinforcement learning have witnessed a considerable success. However, some problems like the absence of supervising signal in supervised learning are not yet solved. In this paper, a novel approach has been proposed based on the deep deterministic policy gradient algorithm to address such kind of problems, forming the training of an agent supervised by the simulation environment. As verification, our proposed approach is applied to a modified tower defense game, and the experimental results confirmed the effectiveness of our proposed approach. © 2018, Curran Associates Inc. All rights reserved.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Logarithm Radon-Fourier transform for high speed small target detection with atmospheric resistance effect

引用

Digital Signal Processing 2025年 165卷

作者： Difeng Sun He Xu Zutang Wu Jin Li Youcao Wu Jianbing Li State Key Laboratory of Complex Electromagnetic Environmental Effects on Electronics and Information System College of Electronic Science and Technology National University of Defense Technology Changsha 410073 China National Key Laboratory of Intense Pulsed Radiation Simulation and Effect Northwest Institute of Nuclear Technology Xi'an 710024 China Academy of Military Science of the People's Liberation Army Beijing 100071 China

Detecting high-speed, small targets in air is challenging due to their complex non-linear motion model and range migration effects. Traditional finite order polynomial models often fail to account for atmospheric drag accurately. This paper introduces a logarithmic relationship between range and time to model this motion, employing the Logarithmic Radon-Fourier Transform (LRFT) for accurately compensating the envelope and phase migration over extended accumulation periods. It offers both the ergodic search method and the fast implementation method for solving LRFT, along with a theoretical analysis of lobe location and amplitude. simulation results indicate that compared with traditional RFT-based methods, LRFT improves accumulation gain, enhances the main lobe's relative amplitude over sidelobes, and offers better detection performance in noisy environments. It provides more accurate recovery of target motion and improved target distinction capabilities. These benefits over traditional methods are especially evident when the product of the attenuation coefficient α , the initial velocity v 0 , and the total coherent accumulation time T M exceeds a specific threshold.

关键词： Fast Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

Privacy sensitive large-margin model for face de-identification 1st

Privacy sensitive large-margin model for face de-identifica...

引用

1st International Conference on Neural Computing for Advanced Applications, NCAA 2020

作者： Guo, Zhiqiang Liu, Huigui Kuang, Zhenzhong Nakashima, Yuta Babaguchi, Noboru Key Laboratory of Complex Systems Modeling and Simulation School of Computer Science and Technology Hangzhou Dianzi University Hangzhou310018 China Osaka University Osaka Japan

ISBN: (纸本)9789811576690

There is an increasing concern of face privacy protection along with the wide application of big media data and social networks due to free online data release. Although some pioneering works obtained some achievements, they are not sufficient enough for sanitizing the sensitive identity information. In this paper, we propose a generative approach to de-identify face images yet preserving the non-sensitive information for data reusability. To ensure a high privacy level, we introduce a large-margin model for the synthesized new identities by keeping a safe distance with both the input identity and existing identities. Besides, we show that our face de-identification operation follows the (Formula Presented)-differential privacy rule which can provide a rigorous privacy notion in theory. We evaluate the proposed approach using the vggface dataset and compare with several state-of-the-art methods. The results show that our approach outperforms previous solutions for effective face privacy protection while preserving the major utilities. © Springer Nature Singapore Pte Ltd. 2020.

关键词： Privacy-preserving techniques

来源：评论

学校读者我要写书评

暂无评论

Chemical and optical properties of aerosols and their interrelationship in winter in the megacity Shanghai of China

引用

Journal of Environmental Sciences 2015年第1期27卷 59-69页

作者： Tingting Han Liping Qiao Min Zhou Yu Qu Jianfei Du Xingang Liu Shengrong Lou Changhong Chen Hongli Wang Fang Zhang Qing Yu Qiong Wu State Key Laboratory of Water Environment Simulation School of Environment Beijing Normal University State Environmental Protection Key Laboratory of Cause and Prevention of Urban Air Pollution Complex Shanghai Academy of Environmental Sciences State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry Institute of Atmospheric PhysicsChinese Academy of Sciences Shanghai Meteorological Service College of Global Change and Earth System Science Beijing Normal University

A field campaign on air quality was carried out in Shanghai in winter of 2012. The concentrations of NO, NO2, NOx, SO2, CO, and PM2.5increased during haze formation. The average masses of SO4^2-, NO3^-and NH4^＋were 10.3, 11.7 and 6.7 μg/m^3 during the haze episodes, which exceeded the average（9.2, 7.9, and 3.4 μg/m3） of these components in the non-haze days. The mean values for the aerosol scattering coefficient（b sp）, aerosol absorption coefficient（b ap） and single scattering albedo（SSA） were 288.7, 27.7 and0.91 Mm-1, respectively. A bi-peak distribution was observed for the mass concentrations of CO, NO, NO2, and NOx. More sulfate was produced during daytime than that in the evening due to photochemical reactions. The mass concentration of NH4＋achieved a small peak at noontime. NO3-showed lower concentrations in the afternoon and higher concentrations in the early morning. There were obvious bi-peak diurnal patterns for b sp and b ap as well as SSA. b sp and b ap showed a positive correlation with PM2.5mass concentration.（NH4）2SO4, NH4NO3, organic mass, elemental carbon and coarse mass accounted for 21.7%, 19.3%, 31.0%, 9.3% and 12.3% of the total extinction coefficient during non-haze days, and 25.6%, 24.3%, 30.1%, 8.1% and 8.2% during hazy days. Organic matter was the largest contributor to light extinction. The contribution proportions of ammonium sulfate and ammonium nitrate to light extinction were significantly higher during the hazy time than during the non-haze days.

关键词： Optical properties Diurnal variations Chemical apportionment Shanghai

来源：评论

学校读者我要写书评

暂无评论

Tightly-coupled integration of GPS/INS and simultaneous localisation and mapping

Tightly-coupled integration of GPS/INS and simultaneous loca...

引用

Australasian Conference on Robotics and Automation 2016, ACRA 2016

作者： Kim, Jonghyuk Cheng, Jiantong Guivant, Jose Australian National University Australia Key Laboratory of Complex Aviation System Simulation Beijing China School of Mechanical and Manufacturing Engineering UNSW Australia

ISBN: (纸本)9781634396080

This paper presents a tightly integrated navigation system combining global positioning system (GPS) and inertial-based simultaneous localization and mapping (SLAM) for UAV platforms. GPS raw measurements, called pseudorange and pseudorange rate, are directly fused to an inertial SLAM filter. A compressed form of unscented filtering is implemented by partitioning the map into a local and global one. The performance of the proposed method is analysed using a high-fidelity 6-degrees-of-freedom simulator, demonstrating accurate and robust navigation even under a single satellite observation. The information gain of bearing and elevation angles is further analysed offering effective sensing strategies. © 2018 Australasian Robotics and Automation Association. All rights reserved.

关键词： Global positioning system

来源：评论

学校读者我要写书评

暂无评论

Autonomy Evaluation of Unmanned systems Based on Task Models

引用

Machine Intelligence Research 2024年第5期21卷 815-830页

作者： Yi Zou Zehao Ni Xun Lei Chi Zhang Key Laboratory of Complex Aviation Simulation System Beijing100076China State Key Laboratory of Multimodal Artificial Intelligence Systems(MAIS) Institute of AutomationChinese Academy of Sciences(CASIA)Beijing100190China Hong Kong Institute of Science&Innovation Chinese Academy of Sciences(HKISI CAS)Hong Kong999077China

In this study,relevant work on autonomy evaluation(AE)in recent years was comprehensively reviewed and classified from the perspective of task models,and a closed-loop task models based theoretical framework for AE was *** main contributions of this study are as follows.1)A taxonomy for AE based on task models was introduced to classify current theories,methods and standards.2)The limitations of the current autonomous evaluation methods were addressed to provide a theoretical framework for quantitative evaluation based on task models,and evaluation metrics for each stage were proposed based on the AE theoretical framework.3)Qualitative analyses of the superiority of the proposed AE framework based on the closed-loop task models were *** study attempts to provide a reference for researchers and engineers in the autonomous unmanned systems field and inspire future development of AE.

关键词： Unmanned system(UMS) autonomy evaluation task models closed-loop task models kill chain

来源：评论

学校读者我要写书评

暂无评论

A Unified Approach to Construct Generalized B-Splines for Isogeometric Applications

引用

Journal of systems Science & complexity 2017年第4期30卷 983-998页

作者： XU Gang SUN Ningning XU Jinlan HUI Kin-Chuen WANG Guozhao Department of Computer Science Hangzhou Dianzi UniversityHangzhou 310018China Key Laboratory of Complex Systems Modeling and Simulation Ministry of EducationHangzhou 310018China Department of Mechanical and Automation Engineering The Chinese University of Hong KongHong KongChina Department of Mathematics Zhejiang UniversityHangzhou 310027China

Abstract Generalized B-splines have been employed as geometric modeling and numerical simu- lation tools for isogeometric analysis （IGA for short）. However, the previous models used in IGA, such as trigonometric generalized B-splines or hyperbolic generalized B-splines, are not the unified mathematical representation of conics and polynomial parametric curves/surfaces. In this paper, a unified approach to construct the generalized non-uniform B-splines over the space spanned by {α（t）,β（t）,ξ（t）, η（t）, 1, t,……. , tn-4} is proposed, and the corresponding isogeometric analysis framework for PDE solving is also studied. Compared with the NURBS-IGA method, the proposed frameworks have several advantages such as high accuracy, easy-to-compute derivatives and integrals due to the non-rational form. Furthermore, with the proposed spline models, isogeometric analysis can be performed on the computational domain bounded by transcendental curves/surfaces, such as the involute of circle, the helix/helicoid, the catenary/catenoid and the cycloid. Several numerical examples for isogeometrie heat conduction problems are presented to show the effectiveness of the proposed methods.

关键词： Generalized B-splines heat conduction isogeometric analysis unified construction.

来源：评论

学校读者我要写书评

暂无评论

Radar Signal Modulation Recognition Based on Bispectrum Features and Deep learning 2

Radar Signal Modulation Recognition Based on Bispectrum Feat...

引用

2nd International Conference on Computer Engineering and Application, ICCEA 2021

作者： Dong, Zeyu Lv, Fengrong Wan, Tao Jiang, Kaili Fang, Xueli Zhang, Lei University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu China Key Laboratory of Complex Aviation Simulation System Beijing China

ISBN: (纸本)9781665426169

Signal bispectral transformation can not only suppress the influence of Gaussian white noise on signal modulation recognition, but also retain the signal amplitude and phase information. It is also used to extract the non-linear characteristics. Compared with other high-order spectra, bispectrum has a simple processing flow. However, the direct use of all bispectrum as signal features will lead to two-dimensional template matching, causing lots of calculations. Converting two-dimensional bispectrum into one-dimensional sequence, for example, extracting slice information of bispectrum, or using integral bispectrum apparently reduce the amount of data to be processed while retaining part of the bispectrum information. We input the extracted bispectral transformation of radar signals into the neural network to realize modulation recognition. The simulations validate our conclusions that our proposed methods still have a high recognition probability while SNR is low. © 2021 IEEE.

关键词： Template matching

来源：评论

学校读者我要写书评

暂无评论

Radar Signal Detection Based on Ambiguity Function and Deep Learning 2

Radar Signal Detection Based on Ambiguity Function and Deep ...

引用

2nd International Conference on Computer Engineering and Application, ICCEA 2021

作者： Lv, Fengrong Dong, Zeyu Wan, Tao Jiang, Kaili Fang, Xueli Zhang, Lei University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu China Key Laboratory of Complex Aviation Simulation System Beijing China

ISBN: (纸本)9781665426169

The ambiguity function describes the joint characteristics of the signal in the time domain and the frequency domain, and is usually used for signal recognition. The time domain and frequency domain distribution of noise is more spurious than the signal. Its ambiguity function can also reflect its characteristics. However, the direct use of the ambiguity function as the signal feature will increase the amount of calculation of the detection system, so slice it, which reduces the burden on the system for detecting signal information while preserving the information. At the same time, the radar signal detection based on neural network can avoid the influence of artificially setting the threshold on the detection performance of the system. So, we propose an algorithm for inputting the extracted fuzzy function slice sequence into a neural network to realize signal detection. simulation proves that when the signal-To-noise ratio of the received signal is very low, our proposed method still has a high detection probability. © 2021 IEEE.

关键词： Signal detection

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised Open-set Recognition of Radar Active Jamming

Semi-supervised Open-set Recognition of Radar Active Jamming

引用

2021 CIE International Conference on Radar, Radar 2021

作者： Li, Hao Fang, Xueli Zhang, Lei Kang, Hui Zhang, Wei School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China Key Laboratory of Complex Aviation Simulation System Beijing China

ISBN: (纸本)9781665498142

With the emergence of various kinds of complex radar jamming, radar jamming intelligent recognition technology has increasingly become the key to electronic counter-counter measures. To achieve radar jamming open-set recognition with a small number of tagged samples, Generative adversarial network-based pseudolabeled convolutional neural network (GAN-PL-CNN) model is proposed in this paper. Firstly, radar jamming is modelled, and the radar jamming datasets with known and unknown types are established after time-frequency transformation and image processing. Then, the adaptive pseudolabeling method is used to achieve semi-supervised recognition of radar jamming by combining the learning capability of convolutional neural networks and the characteristics of pseudolabeling to increase labeled samples intelligently. Finally, generative adversarial networks are introduced to enhance the recognition ability of the model and the discrimination ability of the unknown type through the game between the generator and the discriminator in the iterative process. Experimental results show that the GAN-PL-CNN model achieves 89.7%-95.1% for known types and 30.7%-86.8% for unknown types at 3%-6% labeling rate. As the number of labeled samples increases, the recognition effect becomes better and has a greater ability to exclude unknown type. © 2021 IEEE.

关键词： convolutional neural network open-set recognition pseudo-label Radar active jamming

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：