检索结果-南通市图书馆

Fractional partial differential equation denoising models for texture image

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2014年第7期57卷 184-202页

作者： PU YiFei SIARRY Patrick ZHOU JiLiu LIU YiGuang ZHANG Ni HUANG Guo LIU YiZhi School of Computer Science and Technology Sichuan University Universit de Paris 12 (lissi E.A.3956) 61 av.du Gnral de Gaulle94010 CRETEIL CedexFrance Library of Sichuan University Sichuan University Computer Science College Leshan Normal University Wu Yuzhang Honors College of Sichuan University

In this paper,a set of fractional partial differential equations based on fractional total variation and fractional steepest descent approach are proposed to address the problem of traditional drawbacks of PM and ROF multi-scale denoising for texture *** extending Green,Gauss,Stokes and Euler-Lagrange formulas to fractional field,we can find that the integer formulas are just their special case of fractional *** order to improve the denoising capability,we proposed 4 fractional partial differential equation based multiscale denoising models,and then discussed their stabilities and convergence *** deduction and experimental evaluation demonstrate the stability and astringency of fractional steepest descent approach,and fractional nonlinearly multi-scale denoising capability and best value of parameters are discussed *** experiments results prove that the ability for preserving high-frequency edge and complex texture information of the proposed denoising models are obviously superior to traditional integral based algorithms,especially for texture detail rich images.

关键词： fractional Green formula fractional Euler-Lagrange equation fractional steepest descent approach fractional extreme points fractional total variation fractional differential mask.

Characterization of a human brain cortical surface mesh using discrete curvature classification and digital elevation model

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Biomedical Science and Engineering 2012年第3期5卷 133-140页

作者： Kamel Aloui Amine Nait-Ali Mohamed Saber Naceur lissi EA LaboratoryUniversite Paris-Est Creteil(UPEC)CreteilFrance LTSIRS Laboratory National Engineering SchoolTunisTunisia

In this paper we present a novel approach for brain surfacec characterization based on convexity and concavity analysis of cortical surface mesh. Initially, volumetric Magnetic Resonance Images (MRI) data is processed to generate a discrete representation of cortical surface using low-level segmentation tools and Level-Sets method. Afterward, pipeline procedure for brain characterization/labeling is developed. The first characterization method is based on discrete curvature classification. This is consists on estimating curvature information at each vertex in the cortical surface mesh. The second method is based on transforming the brain surface mesh into Digital Elevation Model (DEM), where each vertex is designed by its space coordinates and geometric measures related to a reference surface. In other word, it consists on analyzing the cortical surface as a topological map or an elevation map where the ridge or crest lines represent cortical gyri and valley lines represents sulci. The experimental results have shown the importance of these characterization methods for the detection of significant details related to the cortical surface.

关键词： Brain Curvature DEN Level-Sets MRI Mapping Surface Mesh

CMAC神经网络结构参数及其结构优化的研究

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

西北工业大学学报 2008年第6期26卷 732-737页

作者：于薇薇闫杰 C．Sabourin K．Madani 西北工业大学航天学院飞行控制与仿真研究所陕西西安710072 巴黎十二大学lissi实验室

CMAC神经网络具有学习算法简单、收敛速度快、局域泛化等优点,被广泛应用于机器人控制、信号处理、模式识别以及自适用控制等领域。但是网络的训练过程需要大量的存储单元,最优结构参数的选取是CMAC网络设计中一个重要问题。文中通过对... 详细信息

CMAC神经网络具有学习算法简单、收敛速度快、局域泛化等优点,被广泛应用于机器人控制、信号处理、模式识别以及自适用控制等领域。但是网络的训练过程需要大量的存储单元,最优结构参数的选取是CMAC网络设计中一个重要问题。文中通过对函数逼近问题的研究,说明了量化精度和泛化参数如何影响网络对函数的逼近质量。仿真结果表明,通过对结构参数的调整,可以达到最小的逼近误差。而通过对网络结构的优化不但可以节约网络的训练时间而且可以大幅度减少存储单元的数量。

关键词： CMAC神经网络泛化参数结构优化函数逼近

基于融合特征以及卷积神经网络的环境声音分类系统研究

在线全文

学校读者我要写书评

暂无评论

西北工业大学学报 2020年第1期38卷 162-169页

作者：张科苏雨王靖宇王霰宇张彦华航天飞行动力学技术重点实验室陕西西安710072 西北工业大学航天学院陕西西安710072 Signals Imagesand Intelligent Systems Laboratory(LISSI/EA 3956)University Paris-Est CreteilSenart-FB Institute of Technology36-37 rue Charpak77127 LieusaintFrance

环境声音识别系统主要基于深度神经网络以及种类繁多的听觉特征对环境声音进行分类识别。分析基于深度神经网络的环境分类任务中,哪种听觉特征更适合环境声音识别系统十分必要。选择了基于2个广泛使用的滤波器:梅尔和Gammatone滤波器组... 详细信息

环境声音识别系统主要基于深度神经网络以及种类繁多的听觉特征对环境声音进行分类识别。分析基于深度神经网络的环境分类任务中,哪种听觉特征更适合环境声音识别系统十分必要。选择了基于2个广泛使用的滤波器:梅尔和Gammatone滤波器组提取的3种声音特征。随后,提出了一个MFCC和GFCC融合的特征MGCC。最后采用文中提出的深度卷积神经网络来验证哪种特征更适合于环境声音的分类识别。实验结果表明,在基于神经网络的环境声音分类系统中,信号处理特征比频谱图特征的效果好,其中,MGCC特征具有比其他特征更好的性能。最后,用文中提出的MCC-CNN模型与其他环境声音分类模型在UrbanSound 8K数据集上进行了对比。实验结果表明,所提模型分类精度最好。

关键词：环境声音特征融合声音分类卷积神经网络

Generating Efficient Solutions in Bilevel Multi-Objective Programming Problems

在线全文

学校读者我要写书评

暂无评论

American Journal of Operations Research 2013年第2期3卷 289-298页

作者： Calice Olivier Pieume Patrice Marcotte Laure Pauline Fotso Patrick Siarry Department of Computer Science and Operations Research University of Montreal Montreal Canada Department of Computer Sciences Faculty of Science University of Yaoundé I Yaoundé Cameroon Department of Mathematics Faculty of Science University of Yaoundé I Yaoundé Cameroon lissi Faculty of Sciences and Technology University of Paris XII Val-de-Marne Créteil France

In this paper, we address bilevel multi-objective programming problems (BMPP) in which the decision maker at each level has multiple objective functions conflicting with each other. Given a BMPP, we show how to construct two artificial multiobjective programming problems such that any point that is efficient for both the two problems is an efficient solution of the BMPP. Some necessary and sufficient conditions for which the obtained result is applicable are provided. A complete procedure of the implementation of an algorithm for generating efficient solutions for the linear case of BMPP is presented. A numerical example is provided to illustrate how the algorithm operates.

关键词： Multi-Objective Programming Bilevel Programming Efficient Solution Efficient Edge Hierarchical Systems

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

双足机器人在动态环境中步态规划研究

计算机仿真 2009年第4期26卷 218-222页

作者：于薇薇 C.Sabourin K.Madani 闫杰西北工业大学航天学院飞行控制与仿真实验室陕西西安710072 巴黎十二大lissi实验室IUT de Senart 77127 LieusaintFrance

在双足机器人跨越迎面而来的动态障碍物的问题中,由于障碍物的高度,和速度是不可预测等因素,机器人的迈步步长和迈步高度决定了其能否实现成功跨越。介绍一种双足机器人步态规划方法,应用模糊Q学习算法对迈步高度进行学习,将迈步的起始... 详细信息

在双足机器人跨越迎面而来的动态障碍物的问题中,由于障碍物的高度,和速度是不可预测等因素,机器人的迈步步长和迈步高度决定了其能否实现成功跨越。介绍一种双足机器人步态规划方法,应用模糊Q学习算法对迈步高度进行学习,将迈步的起始点、落点和迈步高度作为特征点,利用三次样条对特征点进行插值得到摆动腿运动轨迹,最后通过摆动角间的几何关系得出各关节处摆动角的变化规律,控制机器人跨越动态障碍物。仿真结果表明,通过进行的步态规划,机器人可以成功跨越动态障碍物,并且各关节处的摆动角变化曲线平缓无畸变。

关键词：双足机器人步态规划模糊学习算法三次样条插值

基于Adams和Matlab联合仿真的双足机器人越障研究

在线全文

学校读者我要写书评

暂无评论

计算机测量与控制 2008年第11期16卷 1741-1743,1769页

作者：于薇薇 C. Sabourin K. Madani 闫杰西北工业大学航天学院陕西西安710072 巴黎十二大lissi实验室 france77127

与轮行机器人相比,双足机器人具有更灵活的机械结构,具有跨越静态或动态障碍物的能力,使其可以在更复杂的环境中工作;以往的双足机器人路径规划控制策略只能解决静止或以可预测速度运动的障碍物的越障问题,提出了一种基于模糊Q学习算法... 详细信息

与轮行机器人相比,双足机器人具有更灵活的机械结构,具有跨越静态或动态障碍物的能力,使其可以在更复杂的环境中工作;以往的双足机器人路径规划控制策略只能解决静止或以可预测速度运动的障碍物的越障问题,提出了一种基于模糊Q学习算法的路径规划策略,在Adams软件中建立机器人的三维虚拟样机模型,在Matlab软件中设计控制器,进行联合仿真;仿真结果表明所设计的控制策略可以有效地克服机器人在线学习时间长的问题,并且可以成功跨越速度不可预测的运动障碍物,有很好的鲁棒性。

关键词：双足机器人路径规划模糊Q学习算法 Adams软件

在线全文

学校读者我要写书评

暂无评论

双足机器人跨越动态障碍物在线控制系统设计

计算机测量与控制 2008年第10期16卷 1441-1443,1447页

作者：于薇薇 Sabourin C Madani K 闫杰西北工业大学航天学院飞行控制与仿真实验室陕西西安710072 巴黎十二大lissi实验室IUT deSenart

在双足机器人跨越动态障碍物的在线控制问题中,脚步规划和步态控制的学习时间是关键问题;提出了一种将机器人的步态控制和脚步规划分别独立设计的控制策略;步态控制目的是产生关节点轨迹并控制对理想轨迹的跟踪,考虑到双足机器人关节点... 详细信息

在双足机器人跨越动态障碍物的在线控制问题中,脚步规划和步态控制的学习时间是关键问题;提出了一种将机器人的步态控制和脚步规划分别独立设计的控制策略;步态控制目的是产生关节点轨迹并控制对理想轨迹的跟踪,考虑到双足机器人关节点轨迹的不连续性,应用小脑模型连接控制CMAC记忆特征步态的关节点轨迹;脚步规划的控制目标是通过对环境的视觉感知预测机器人的运动路径,算法是基于无需对动态环境精确建模的模糊Q学习算法;仿真结果表明该控制策略的可行性,并且可以有效缩短在线学习时间。

关键词：步态脚步规划小脑模型连接控制器模糊Q学习算法

Longitudinal analysis of quantitative biomarkers using projection-resolved OCT angiography in retinal vein occlusion

在线全文

学校读者我要写书评

暂无评论

Annals of Eye Science 2021年第2期6卷 19-31页

作者： Alexandra Miere Donato Colantuono Camille Jung Agnès Glacet-Bernard Severine Becuwe Eric Petit Eric Souied Department of Ophthalmology Centre Hospitalier Intercommunal de CréteilUniversitéParis EstCréteilFrance Laboratory of Images Signals and Intelligent Systems Laboratory(LISSI/EA 3956)University Paris-EstCréteilFrance Clinical Research Center GRC Maculaand Biological Ressources CenterCentre Hospitalier Intercommunal de CréteilCréteilFrance UNIACT NeurospinISVFJCEAGif-sur-YvetteFrance

Background:To evaluate a fully automated vascular density(VD),skeletal density(SD)and fractal dimension(FD)method for the longitudinal analysis of retinal vein occlusion(RVO)eyes using projection-resolved optical coherence tomography angiography(OCTA)images and to evaluate the association between these quantitative variables and the visual prognosis in RVO ***:Retrospective longitudinal observational case *** presenting with RVO to Creteil University Eye Clinic between October 2014 and December 2018 and healthy controls were retrospectively *** 1 consisted of central RVO(CRVO)eyes,group 2 consisted of eyes with branch RVO(BRVO)and group 3 of healthy control *** acquisitions(AngioVue RTVue XR Avanti,Optovue,Inc.,Freemont,CA)were performed at baseline and last follow up ***,SD,and FD analysis were computed on OCTA superficial and deep vascular complex(SVC,DVC)images at baseline and final follow up using an automated *** regression was performed to find if and which variable(VD,SD,FD)was predictive for the visual ***:Forty-one eyes,of which 21 consecutive eyes of 20 RVO patients(13 CRVO in group 1,8 BRVO in group 2),and 20 eyes of 20 healthy controls were *** the level of SVC,VD and FD were significantly lower in RVO eyes compared to controls(P<0.0001 and P=0.0008 respectively).Best-corrected visual acuity(BCVA)at last follow-up visit was associated with baseline VD(P=0.013),FD(P=0.016),and SD(P=0.01)at the level of the SVC,as well as with baseline FD at the DVC level(P=0.046).Conclusions:Baseline VD,SD,and FD are associated with the visual outcome in RVO *** parameters seem valuable biomarkers and may help improve the evaluation and management of RVO patients.

关键词： Retinal vein occlusion(RVO) vascular density(VD) optical coherence tomography angiography(OCTA) fractal dimension(FD) image analysis