摘要
模式识别结果的好坏往往取决于对识别对象本身特征的提取。对于一类对象通常已经存在若干成形的提取方法,要想达到较好的识别分类效果就需要在特征提取方法的层面上获取更具代表性的特征表达。以字符识别所用的特征提取方法为例,研究特征提取方法的分析评价。用5种常用的特征提取方法及其组合对标准字模进行特征提取,分别使用最小欧氏距离、欧氏距离总和及欧氏距离方差3种测度在4个不同维度空间内对提取方法进行比较分析。最后,综合使用3种测度值去评价特征提取方法。
Whether the result of pattern recognition is good depends on the feature extraction of the objects being recognized. There usually exit some formed extraction-methods for the objects of a certain kind. So if better result of recognition and classification is needed, more representative features should be extracted for the feature-extraction methods. In this paper, we take the feature extraction-method used in character recognition as an example to do the research on analysis and evaluation of feature extraction-methods. We use five common types of feature extractionmethods and their combinations to extract the features of the standard characters. Then we use three types of measures to compare and analyze the extraction-methods in four different feature spaces respectively, which are minimum value of Euclidean distance, sum of Euclidean distance and variance of Euclidean distance. At last, we synthetically use the three kinds of measures to evaluate the feature extraction-methods.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2010年第4期898-903,共6页
Chinese Journal of Scientific Instrument
基金
北京交通大学科技基金(2006XM028)资助项目
关键词
模式识别
特征提取
欧氏距离
特征空间
综合测度
pattern recognition
feature extraction
Euclidean distance
feature space
compositive measurement
