检索结果-内蒙古大学图书馆

International Conference on Document Analysis and recognition

作者： Zhihua Yang Lihua Yang C.Y. Suen School of Mathematical Sciences South China Normal University Guangzhou China School of Mathematics and Computing Science Sun Yat-sen University Guangzhou China Center for Pattern Recognition and Machine Intelligence Concordia University Montreal Canada

Chinese fonts are recognized by a new method based on empirical mode decomposition. Five basic strokes have been selected to characterize the features of Chinese fonts. Based on them, stroke feature sequences of a given text block are calculated. Once decomposed by EMD, the first two intrinsic mode functions corresponding to each stroke feature sequence are used to calculate the stroke energy of all the five basic strokes. These energies are combined with the five averages of the residues to produce a ten-dimensional feature vector. Finally, the minimum distance classifier is used to recognize the fonts. Experiments show encouraging recognition rates.

关键词： Text recognition Feature extraction Sun pattern recognition Text analysis Computational complexity Character recognition Mathematics machine intelligence Image recognition

来源：评论

学校读者我要写书评

暂无评论

Low-level cursive word representation based on geometric decomposition

引用

4th International Conference on machine Learning and Data Mining in pattern recognition, MLDM 2005

作者： Dong, Jian-Xiong Krzyzak, Adam Suen, Ching Y. Ponson, Dominique IMDS Software 75 rue Queen Montréal Que. H3C 2N6 Canada Center for Pattern-Recognition and Machine Intelligence Concordia University Montréal Que. H3G 1M8 Canada Department of Computer Science and Software Engineering Concordia University 1455 de Maisonneuve Blvd. W. Montréal Que. H3G 1M8 Canada

ISBN: (纸本)3540269231

An efficient low-level word image representation plays a crucial role in general cursive word recognition. This paper proposes a novel representation scheme, where a word image can be represented as two sequences of feature vectors in two independent channels, which are extracted from vertical peak points on the upper external contour and at vertical minima on the lower external contour, respectively. A data-driven method based on support vector machine is applied to prune and group those extreme points. Our experimental results look promising and have indicated the potential of this low-level representation for complete cursive handwriting recognition. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Image analysis

来源：评论

学校读者我要写书评

暂无评论

Automatic segmentation of unconstrained handwritten numeral strings

Automatic segmentation of unconstrained handwritten numeral ...

引用

International Workshop on Frontiers in Handwriting recognition

作者： J. Sadri C.Y. Suen T.D. Bui Center for Pattern Recognition and Machine Intelligence (CENPARMI) Concordia University Montreal QUE Canada

A new method of segmenting unconstrained handwritten numeral strings is proposed. It is based on the extracting of foreground and background features. In order to find foreground features for the first time an algorithm based on skeleton tracing is introduced. The skeleton of each connected component is traversed in clockwise and anti-clockwise directions, and intersection points which are visited in each traversal, are mapped on the outer contour to form foreground feature points. In order to find background features, another new algorithm is proposed. Considering vertical projections of top and bottom profiles, two background skeletons are found. After processing these two background skeletons, background feature points are extracted. Background and foreground feature points are assigned together to construct candidate segmentation paths. Finally each segmentation path is evaluated based on the properties of its left and right connected components. Our method can provide a list of good segmentation hypotheses for segmentation-based recognition systems. The NIST SD19 database (handwritten numeral strings) is used for evaluating of the method, and experiments show a very promising result.

关键词： Skeleton Image segmentation Feature extraction Handwriting recognition Character recognition NIST Spatial databases Optical character recognition software Pixel pattern recognition

来源：评论

学校读者我要写书评

暂无评论

A segmentation method for touching italic characters

A segmentation method for touching italic characters

引用

International Conference on pattern recognition

作者： Yun Li S. Naoi M. Cheriet C.Y. Suen Center of Pattern Recognition and Machine Intelligence Concordia University Montreal QUE Canada Document Processing Laboratory Fujitsu Laboratories Limited Kawasaki Japan Laboratory for Imagery Vision and Artificial Intelligence Ecole de Technologie Supérieure University of Quebec Montreal QUE Canada

ISBN: (纸本)0769521282

Segmentation is an essential part of a recognition system. It is difficult to handle touching characters, especially for italic fonts. We present a method to achieve the accurate segmentation of touching italic characters. It is free of slant correction, so extra noises will not be introduced. We use slant projection and contour analysis to find the segmentation points. Then the shortest path approach is adopted for accurately locating the cut path of each candidate segmentation point. Based on dynamic programming, we can find the best segmentation result from those cut paths. By this method, we can improve the recognition rate on italic fonts.

关键词： Image segmentation Histograms pattern recognition Image recognition Dynamic programming Character recognition machine intelligence Laboratories Artificial intelligence Optical character recognition software

来源：评论

学校读者我要写书评

暂无评论

Image recognition Based on Nonlinear Wavelet Approximation

引用

International Journal of Wavelets, Multiresolution and Information Processing 2003年第2期1卷 151-161页

作者： L. H. Yang T. D. Bui C. Y. Suen Department of Scientific Computing and Computer Application Zhongshan University Guangzhou 510275 P. R. China Center for Pattern Recognition and Machine Intelligence Concordia University Montreal Canada H3G 1M8 Canada

This paper presents a novel approach to recognize images based on nonlinear wavelet approximation. The mathematical theory on nonlinear wavelet approximation is introduced, which shows that nonlinear approximation contains much more information of the original image than linear approximation. Based on this theory, a scheme to obtain the basic information of images with less data is provided. Experiments on face recognition produce effective matching rates.

关键词： Wavelet basis nonlinear approximation image recognition

来源：评论

学校读者我要写书评

暂无评论

Word segmentation of printed text lines based on gap clustering and special symbol detection

Word segmentation of printed text lines based on gap cluster...

引用

作者： Kim, Soo H. Jeong, Chang B. Kwag, Hee K. Suen, Ching Y. Department of Computer Science Chonnam National University Korea Republic of Center for Artificial Intelligence Research KAIST Korea Republic of Center for Pattern Recognition and Machine Intelligence Concordia University Canada

This paper proposes a word segmentation method for machine-printed text lines. It utilizes gaps and special symbols as delimiters between words. A gap clustering technique is used to identify the gaps between words regardless of the gap-size variations among different document images. Next a special symbol detection technique is applied to find two types of special symbols lying between words. An experiment with 1.675 text lines in 100 different English and Korean documents shows that the proposed method achieves a high accuracy of word segmentation. © 2002 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Word segmentation of printed text lines based on gap clustering and special symbol detection

Word segmentation of printed text lines based on gap cluster...

引用

16th International Conference on pattern recognition, ICPR 2002

ISBN: (纸本)0769516963

This paper proposes a word segmentation method for machine-printed text lines. It utilizes gaps and special symbols as delimiters between words. A gap clustering technique is used to identify the gaps between words regardless of the gap-size variations among different document images. Next a special symbol detection technique is applied to find two types of special symbols lying between words. An experiment with 1,675 text lines in 100 different English and Korean documents shows that the proposed method achieves a high accuracy of word segmentation. © 2002 IEEE.

关键词： Signal detection

来源：评论

学校读者我要写书评

暂无评论

An application of nonlinear wavelet approximation to face recognition

An application of nonlinear wavelet approximation to face re...

引用

作者： Yang, L.H. Bui, T.D. Suen, C.Y. Department of Scientific Computing and Computer Application Zhongshan University Guangzhou 510275 China Center for Pattern Recognition and Machine Intelligence Concordia University Montreal H3G 1M8 Canada

In this paper, we present an application of nonlinear wavelet approximation to recognize faces. The advantages of nonlinear wavelet approximation are compared with linear wavelet approximation. An efficient scheme in applying nonlinear wavelet transform to face recognition is presented. Experiments using the Yale Face Library show very encouraging recognition results and that this method has the desirable property of being insensitive to facial expressions. © 2002 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Word segmentation of printed text lines based on gap clustering and special symbol detection

Word segmentation of printed text lines based on gap cluster...

引用

International Conference on pattern recognition

作者： S.H. Kim C.B. Jeong H.K. Kwag C.Y. Suen Department of Computer Science Chonnam National University South Korea Center for Artificial Intelligence Research KAIST South Korea Center for Pattern Recognition and Machine Intelligence Concordia University Canada

This paper proposes a word segmentation method for machine-printed text lines. It utilizes gaps and special symbols as delimiters between words. A gap clustering technique is used to identify the gaps between words regardless of the gap-size variations among different document images. Next a special symbol detection technique is applied to find two types of special symbols lying between words. An experiment with 1,675 text lines in 100 different English and Korean documents shows that the proposed method achieves a high accuracy of word segmentation.

关键词： Image segmentation Optical character recognition software White spaces Size measurement Computer science Artificial intelligence pattern recognition machine intelligence Optical devices Document image processing

来源：评论

学校读者我要写书评

暂无评论

An application of nonlinear wavelet approximation to face recognition

An application of nonlinear wavelet approximation to face re...

引用

International Conference on pattern recognition

作者： L.H. Yang T.D. Bui C.Y. Suen Center for Pattern Recognition and Machine Intelligence Concordia University Montreal Canada Department of Scientific Computing and Computer Application Zhongshan University Guangzhou China

ISBN: (纸本)076951695X

In this paper we present an application of the nonlinear wavelet approximation to recognize faces. The advantages of the nonlinear wavelet approximation are compared with its linear counterpart. An efficient scheme in applying the nonlinear wavelet transform to face recognition is presented. Experiments using the Yale Face Library show very encouraging recognition results and demonstrate that this method has the desirable property of being insensitive to facial expressions.

关键词： Face recognition Image reconstruction Wavelet transforms Application software pattern recognition Databases Linear approximation Face detection Humans Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：