期刊文献+
共找到4篇文章
< 1 >
每页显示 20 50 100
Kinect骨骼数据驱动的人体动作二维特征融合与动作识别 被引量:3
1
作者 张成权 唐家康 汪志峰 《安庆师范大学学报(自然科学版)》 2020年第1期77-83,共7页
针对在高噪声环境中人体动作识别存在准确度和稳定性不高的问题,本文采用二维空间特征融合的方法,提出一种基于Kinect骨骼数据的人体动作识别算法。从人体三视图的投影来提取运动特征,可以消除人体自遮挡的影响。针对人体复杂动作,算法... 针对在高噪声环境中人体动作识别存在准确度和稳定性不高的问题,本文采用二维空间特征融合的方法,提出一种基于Kinect骨骼数据的人体动作识别算法。从人体三视图的投影来提取运动特征,可以消除人体自遮挡的影响。针对人体复杂动作,算法采用分层策略。利用Kinect获得的骨骼关节点坐标,根据人体三视图投影提取二维空间的人体关节角特征,并运用支持向量机(SVM)方法对动作进行粗分类;进一步提取二维投影平面内的关节位置矢量、角速度和加速度特征,运用隐马尔可夫模型(HMM)的方法对动作进行细分类。利用本文方法对公开数据集MSR Action 3D实验,平均识别率达93.37%,实验结果表明,该方法准确性较高,鲁棒性较强。 展开更多
关键词 KINECT 动作识别 骨骼数据 二维空间特征
下载PDF
近30年北京市ISP-LST空间特征及其变化 被引量:15
2
作者 于琛 胡德勇 +3 位作者 曹诗颂 张旸 张亚妮 段欣 《地理研究》 CSSCI CSCD 北大核心 2019年第9期2346-2356,共11页
本文聚焦长时序地表的不透水与温度特征,利用Landsat影像数据,获取1991-2015年北京市的不透水地表盖度(Impervious Surface Percentage, ISP)与地表温度(Land Surface Temperature, LST)数据,构建不透水地表盖度-地表温度(ISP-LST)二维... 本文聚焦长时序地表的不透水与温度特征,利用Landsat影像数据,获取1991-2015年北京市的不透水地表盖度(Impervious Surface Percentage, ISP)与地表温度(Land Surface Temperature, LST)数据,构建不透水地表盖度-地表温度(ISP-LST)二维空间。结合标准差椭圆法,对ISP-LST空间密度分布的聚集特性进行分析,定量化表述各时期的特征与变化。研究发现:①ISP-LST二维空间特征表现为三种类型:弱相关、非完全正相关和显著正相关。②ISP-LST标准差椭圆的方向性和离散性均值为11.26和2.87,空间聚集性良好。随时间推移,高温现象受不透水地表的影响过程趋于复杂化。③ISP-LST聚集区是城市热环境的重要表征,其在各功能区年际增长率为:功能扩展区(2.97%)>核心功能区(1.75%)>发展新区(1.63%)>生态涵养区(0.18%)。聚集区在东南方向增长明显,研究时段内累计增长14.77%。④ISP-LST聚集区的斑块密度及形状复杂度的景观格局变化不大,但斑块连接性随时间推移有所降低。本文研究结果可为缓解城市热岛效应、制定生态环境调控政策提供相应参考。 展开更多
关键词 不透水地表盖度 地表温度 二维空间特征 标准差椭圆 ISP-LST聚集区
原文传递
State Space-Time and Four States of Universe 被引量:5
3
作者 Jinzhong Yan 《Journal of Physical Science and Application》 2013年第2期127-134,共8页
All things in the universe possess a state and characteristics of state, resultantly in presence of space-time, which is perceived by human beings. An outlook of space-time is shaped in human by perceiving the existen... All things in the universe possess a state and characteristics of state, resultantly in presence of space-time, which is perceived by human beings. An outlook of space-time is shaped in human by perceiving the existence and change of objects. The state space is all state characteristics exhibited in objects whilst the state time refers to the duration of an object's state. The time is a spatial property and not an independent dimension. The state space-time is a unity of internal and external space-time. The internal space-time is stemmed from the overall internal forces and internal energies and is a covert space-time. The external space-time refers to a space-time manifested by the external characteristics and movement of an object and is an overt space-time. In physics, there are four kinds of forces and four state space-times: bonding force and three-dimensional space-time; strong interaction of exchangeable n meson and two-dimensional space-time; quark confinement and one-dimensional space-time; and weak interaction and zero-dimensional space-time. The universe is constituted by dissimilar state space-times. Newton space-time is a three-dimensional state space-time; Einstein's theory of relativity is a two-dimensional state space-time. Newton and Einstein were different observers. Temporal and spatial perception of human is dependent upon human's intemal energy and quality. Through Qigong exercises, the human is able to enter the three-dimensional, two-dimensional, one-dimensional and zero-dimensional space-times. The relativity theory of human body will solve the time problems at the interplanetary voyage of astronauts. 展开更多
关键词 State space-time UNIVERSE four states.
下载PDF
A Novel Extension of Kernel Partial Least Squares Regression
4
作者 贾金明 仲伟俊 《Journal of Donghua University(English Edition)》 EI CAS 2009年第4期438-442,共5页
Based on continuum power regression(CPR) method, a novel derivation of kernel partial least squares(named CPR-KPLS) regression is proposed for approximating arbitrary nonlinear functions.Kernel function is used to map... Based on continuum power regression(CPR) method, a novel derivation of kernel partial least squares(named CPR-KPLS) regression is proposed for approximating arbitrary nonlinear functions.Kernel function is used to map the input variables(input space) into a Reproducing Kernel Hilbert Space(so called feature space),where a linear CPR-PLS is constructed based on the projection of explanatory variables to latent variables(components). The linear CPR-PLS in the high-dimensional feature space corresponds to a nonlinear CPR-KPLS in the original input space. This method offers a novel extension for kernel partial least squares regression(KPLS),and some numerical simulation results are presented to illustrate the feasibility of the proposed method. 展开更多
关键词 continuum regression partial least squares kernel function
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部