查询词典 feature vector

The method of simulate wavelet edge multi-scale detection was used to extract feature of characteristic bands successfully,based Gaussian filter and vector differential operator.

提出了基于高斯滤波与矢量微分算子的近似小波变换多尺度杂交玉米种子电泳图谱特征谱带边缘检测算法，成功实现了特征谱带特征参数的提取。

This method mainly studies the change detection based on the judge of temple and threshold value of the land use class. The land use patch is expressed by vector polygon along with raster region. First the spectrum feature databank with histogram, texture and shape features of the patch is formed. Then, the patches are classified and their change types were judged.

其基本思路是在矢栅位置配准的基础上，以矢量多边形和栅格面阵共同表示的土地利用图斑为基本单位，建立土地类别解译标识，在此基础上对所有图斑进行基于图斑的分类、利用检测模板和阈值来判断图斑类型是否发生变化，并对变化图斑进行表示和更新。

Potential support vector machine,as a new wrapper feature selection approach,has been applied to several fields successfully.

GPSFM方法在一定程度上继承了P-SVM的优点，而且还具有特征选择冗余度低、选择速度快和适应能力强的特点，从而使得该方法表现出较之于P-SVM更好的特征选择和分类效果。

One of the most difficulties in document automatic classification is the high dimension of feature space and the sparseness of text representation vector.

文本自动分类问题的最大特点和困难之一是特征空间的高维性和文档表示向量的稀疏性。

This paper study systematically text automatic categorization from three ways including vector model representation, feature selection and classifier training.

讨论了文本表示的整个过程——分词，建立停用词表，特征选择，权重计算，生成向量空间。

Experimental results show that the proposed approach can achieve a better mapping task, while overcoming the shortcoming of iMapper systems that the representation of feature vector over-relies on document collections.

实验表明，OMW方法能够较好的实现映射任务，同时克服了iMapper系统中特征向量的表示过分依赖于文档集合这一不足之处。

This paper study systematically text automatic categorization from these ways including text pretreatment, vector model representation, feature selection and classify methods.

本文从文本预处理，文本的向量模型表示，特征降维方式和分类算法几个部分较系统的研究了文本自动分类。

In order to solve the problem, support vector machine based on weighted feature is proposed in this paper.

为了解决这个问题，本文提出了一种基于特征加权的支持向量回归机。

Animal's welfare can be feed back from their vocalism. We had some experiments to get 35 days old chicken's vocalization in different environment, including normal environment、 light off environment and people's disturbing environment. On the base of Mel Frequency Cepstral Coefficient, We introduction a new algorithm for feature extraction---Subband Mel Spectrum Centroid. We used support vector machine classification model to evaluate the performance of using different sound parameters.

本文通过研究35日龄海兰褐蛋鸡在正常状态、光照与人为干扰等应激下发声信息的变化特征，提出基于子带梅尔频谱质心的优化特征参数，并通过建立支持向量机模型，比较了采用优化特征参数与梅尔频标倒谱系数及其一阶、二阶导数组成的特征参数（MFCC_39）在识别不同应激中的应用效果。

First, a fixed-dimensional feature vector is generated for each protein sequence using the frequency of the hydropathy blocks occurring in the sequence.

首先，利用蛋白质序列中亲水模块的出现频率，每条蛋白质序列被转换为一个特征向量。

Use nationalization degree and the relationship that commercialize degree, condition of calculable market economy plays scope of optimal and state-owned economy.

利用国有化程度与市场化程度的关系，可计算市场经济条件下最佳国有经济规模。

The kinds of extractant also effected desulfurization rate .

萃取剂对脱硫率的影响也很大，乙二胺的萃取效果优于甲醇的。