查询词典 feature

Followed by the decomposition phase is optimal feature subset selection, in which the optimal feature subset acquired the best divergence is obtained according to interclass/intraclass distance of the training samples. This procedure is implemented by a Genetic Algorithm, with each possible feature subset encoded as chromosome. Fitness scores in GA are calculated and evaluated based on Jeffries-Matusita distance of the selected training samples.

然后，对这些线性光谱特征利用遗传算法结合训练样本计算类内/类间距离搜索最优分类子集，其具体染色体编码取可能的特征号，适应度函数基于样本平均Jeffries-Matusita距离计算。

Linear feature extraction in feature space corresponds to nonlinear feature extraction in input space.

特征空间中的线性特征提取对应于输入空间的非线性特征提取。

First,the feature subspaces are formed from different pose images using ICA and PCA.Then the pose image is transformed into its corresponding frontal face image using the transformation matrix predetermined by learning.Based on this foundation,the wavelet transform is applied to face images,and the authors compare the frontal face feature coefficients directly which are obtained by using the transformation matrix with the feature coefficients of the original face images,the recognition rate is improved greatly.

首先利用ICA和PCA提取不同姿态人脸的特征子空间，然后利用通过训练得到的姿态转换矩阵合成其相对应的正面人脸图像，实验表明ICA人脸识别算法要优于PCA人脸识别算法，并在此基础上用小波对人脸图像进行预处理，据姿态转换矩阵得到的正面人脸特征系数直接进行分类比较，识别率得到了很大的提高。

Based on the dual-beam interference principle and white light interference feature, this dissertation designs the SWLI surface 3D topography measuring system and analyzes the principle and characters different from that of traditional monochromatic light interference. According to different descriptions of the white light interference feature, the optimal interference point extracting algorithms are classified, analyzed and compared in detail. In the light of the design rule of improving measuring precision and speed simultaneously, two new types of algorithms, named Feature Matching algorithm and Intensity Contrast Moving Window algorithm, are designed to acquire good results according to practical conditions.

基于双光束干涉原理及白光干涉特性，本文设计了扫描白光干涉法表面三维形貌测量系统，分析了该方法区别于传统单色光干涉测量的测量原理及特点，根据对白光干涉特征不同的描述方式对最佳干涉点提取算法进行了分类、分析与比较，并在提高测量精度与测量速度设计原则的指导下，根据测量系统的实际条件设计了两种新的算法：特征匹配法和对比度移窗法，在实际测量中达到了比较好的测量效果。

This article proposes essential color of color feature and gradation paragenesis matrix of the texture characteristic,as well as the correlation feedback technology carried on together the retrieval.This article uses the correlation feedback technology to adjust the feature coefficient of color and texture and weight value of the block of this two feature.

提出了将颜色特征中的分块主色法，纹理特征中的分块灰度共生矩阵以及相关反馈技术相结合来共同进行检索，并采用相关反馈技术通过更新权值的方法来调整图像颜色、纹理特征系数以及对应于这两种特征的图像各分块的权值。

According to product characteristics and available manufacturing resource, feature was classified and managed by machining equipment. Machining feature base was constructed to realize parametric and modular feature-based NC programming generation.

根据企业的产品特点和可用的制造资源，按加工设备对特征进行分类和管理，建立了加工特征库，实现了基于特征的参数化、模块化数控程序生成。

Segmentation and extraction of the target feature points from the unregistered, noisy data set of human body scanning and measuring are discussed. We have fitted the noise scanned data line with KS principal curve, which is based on the K-means theory. The crossing zero of curvature was used to segment and estimate the position of target feature points. And optimization scheme named variable fuzzy partition is also presented to improve the searching for local feature area.

使用基于K-means思想的KS主曲线算法，对国家人体尺寸测量工作中产生的带噪声扫描线数据集合进行非线性拟合，并依据曲率分布特征提出对目标特征点空间位置进行模糊分区估计的方法；同时提出了变量化模糊分区的优化策略，大幅度地提高了局部特征区域搜索的效率。

After studying the applications of image algebraic feature thoroughly, we found that when algebraic feature is used to face recognition, projective transformation is generally used to compress the dimension of extracted image feature, then classifier is designed to recognize.

对图象代数特征应用作了深入研究之后，发现在应用代数特征进行人脸识别研究时，一般用投影变换来压缩提取的图象特征维数，然后设计分类器进行识别。

The features of automobile panel die have been analyzed and described with discussion about organizing method of the feature information in Teamcenter,a PDM system.Different similarity has been adopted according to different feature property,classified weights have been adopted to solve the problem of the influence of different feature upon the solution hence implementation of cases retrievement.

新闻摘要：有效地利用覆盖件模具开发过程中的经验知识，能够极大地提高模具开发的效率，本文对覆盖件模具的特征进行了分析并对特征进行了描述，对实例特征信息在PDM 系统Teamcenter中的组织进行了探讨，针对不同的属性域采用不同的相似度表示，通过分级权重的方法实现了不同特征对解决方案的影响，实现了实例检索。

It studies taxi passenger's daily travel behavior feature: first, it analyzes temporal distribution of them and proposes the solution of dynamic management of taxis by limiting operation period; second, through investigating the uneven spatial distribution of taxi travel demand and nonlinear correlation of travel density and population density, it reveals the shortcoming of linear matching transportion facility with population number in traditional urban transport criterion, also it questions the limited effect of the licence management for taxis, and it proposes dynamically determing taxi scale according to different traffic zones; third, through exploring the taxi travel patterns in special zones, it reveal Hong Kong citizens travel behavior in Shenzhen, which could analyze the impact of demographic feature to the travel behavior and provide scientific testimony for designing transportion service for the special group; fourth, through accurately grasping the spatial-temporal distribution of taxi travel demand, it could provide innotiative and efficient transportion service at different time and in different place, such as taxi-sharing and flexible shuttle to meet the travel demand; through pervasive data, we could inspect urban expansion and sprawl, monitor the separation of inhatitantion center and employment center and reflect the variation of population centriod and distribution; finally, it reveals how to use digital feature of inhabitant daily travel to detect abnormal condition in the city.

对出租车乘客的日常交通行为特征进行了研究：通过对出租车出行时间分布特征的把握，提出采取限时营运的方法来对出租车进行动态管理的思路；通过分析出租车出行空间分布不均匀的特征以及出行密度与人口密度之间非线性的关系，揭示了传统城市交通规范中按人数来线性配套交通基础设施的缺陷以及牌照管理对于出租车管理的有限效果，提出按照区域动态确定营运出租车规模的思路；通过特定区域出租车出行特征的分析，揭示香港居民在深圳的交通行为特征，从而分析不同人口统计特征对于交通行为的影响，为设计这个特殊群体的交通服务提供科学依据；通过对出租车出行时空分布特征精确的把握，使得我们可以在不同时间、不同地点采用创新、高效的交通服务，如合乘出租车，高级穿梭巴士来满足出行需求；通过普适数据可以监测城市的扩张和蔓延、居住中心和就业中心的分离程度以及反映人口重心与分布的变化；最后探讨了如何利用居民日常出行的数字特征来检测城市异常状态。

Neither the killing of Mr Zarqawi nor any breakthrough on the political front will stop the insurgency and the fratricidal murders in their tracks.

在对危险的南部地区访问时，他斥责什叶派民兵领导人对中央集权的挑衅行为。

In fact,I've got him on the satellite mobile right now.

实际上 我们已接通卫星可视电话了