检索结果-内蒙古大学图书馆

METHODS OF COMBINING MULTIPLE CLASSIFIERS AND THEIR APPLICATIONS TO HANDWRITING recognition

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS 1992年第3期22卷 418-435页

作者： XU, L KRZYZAK, A SUEN, CY Centre for Pattern Recognition and Machine Intelligence Concordia University Montreal QUE Canada

Method of combining the classification powers of several classifiers is regarded as a general problem in various application areas of pattern recognition, and a systematic investigation has been made. Possible solutions to the problem can be divided into three categories according to the levels of information available from the various classifiers. Four approaches are proposed based on different methodologies for solving this problem. One is suitable for combining individual classifiers such as Bayesian, k-NN and various distance classifiers. The other three could be used for combining any kind of individual classifiers. On applying these methods to combine several classifiers for recognizing totally unconstrained handwritten numerals, the experimental results show that the performance of individual classifiers could be improved significantly. For example, on the U.S. zipcode database, the result of 98.9% recognition with 0.90% substitution and 0.2% rejection can be obtained, as well as a high reliability with 95% recognition, 0% substitution and 5% rejection. These results compared favorably to other research groups in Europe, Asia, and North America.

关键词： Handwriting recognition pattern recognition Character recognition Speech recognition Hidden Markov models Remote sensing Classification algorithms Brain modeling Bayesian methods Databases

来源：评论

学校读者我要写书评

暂无评论

METHOD OF COMBINING MULTIPLE EXPERTS FOR THE recognition OF UNCONSTRAINED HANDWRITTEN NUMERALS

引用

IEEE TRANSACTIONS ON pattern ANALYSIS AND machine intelligence 1995年第1期17卷 90-94页

作者： HUANG, YS SUEN, CY Centre for Pattern Recognition and Machine Intelligence Concordia University West Montreal QUE Canada

For pattern recognition, when a single classifier cannot provide a decision which is 100 percent correct, multiple classifiers should be able to achieve higher accuracy. This is because group decisions are generally better than any individual's. Based on this concept, a method called the ''Behavior-Knowledge Space Method'' was developed, which can aggregate the decisions obtained from individual classifiers and derive the best final decisions from the statistical point of view. Experiments on 46,451 samples of unconstrained handwritten numerals have shown that this method achieves very promising performances and outperforms voting, Bayesian, and Dempster-Shafer approaches.

关键词： UNCONSTRAINED HANDWRITING recognition COMBINATION OF MULTIPLE CLASSIFIERS EVIDENCE AGGREGATION BEHAVIOR KNOWLEDGE SPACE KNOWLEDGE MODELING

来源：评论

学校读者我要写书评

暂无评论

BUILDING A NEW GENERATION OF HANDWRITING recognition SYSTEMS

引用

pattern recognition LETTERS 1993年第4期14卷 303-315页

作者： SUEN, CY LEGAULT, R NADAL, C CHERIET, M LAM, L Centre for Pattern Recognition & Machine Intelligence Concordia University Montr&eacute al Canada H3G 1M8

This paper gives an assessment of the current state of the art in handwriting recognition. It summarizes the lessons learned, the difficulties involved, and the challenges ahead. Based on a review of the recent achievements in off-line computer recognition of totally unconstrained handwritten characters, and extensive research, the authors attempt to identify new frontiers for research which may lead to further breakthroughs in this field. They will present some evidences and novel ideas on ways of stretching the limits of handwriting recognition systems aiming at outperforming human beings.

关键词： HANDWRITING recognition FEATURE EXTRACTION IMPROVING HANDWRITING RECOGNIZERS recognition BY EXPERTS

来源：评论

学校读者我要写书评

暂无评论

Multi-class SVM based gradient feature for banknote recognition

Multi-class SVM based gradient feature for banknote recognit...

引用

2017 IEEE International Conference on Industrial Technology, ICIT 2017

作者： Dittimi, Tamarafinide V. Hmood, Ali K. Suen, Ching Y. Centre for Pattern Recognition and Machine Intelligence Concordia University MontrealQC Canada

ISBN: (纸本)9781509053209

Banknote recognition system is the focus of different image processing and pattern recognition research. With the improvement in modern-day banking operations, automated systems for banknote recognition have become pertinent. recognition of banknotes is a challenging task as banknotes can suffer from defects and images get distorted during acquisition, which raises the need for a robust recognition system to mitigate these flaws. This research proposes a new banknote recognition approach that classifies the principal components of the extracted Histogram of Gradient feature vectors using an efficient error correcting output code technique based on a Multi-Class Support Vector machine. The method works on both sides of the bank note and efficiently recognize the denomination based on any side of the bill. The system was implemented using the Nigerian Naira, and for experimental evaluation, additional analysis was conducted using the US Dollar, Canadian Dollar, and Euro banknotes. Finally, the system performance was evaluated based on the recognition rate and processing time. © 2017 IEEE.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

A theoretical analysis of the application of majority voting to pattern recognition 12

A theoretical analysis of the application of majority voting...

引用

12th IAPR International Conference on pattern recognition - Conference B: pattern recognition and Neural Networks, ICPR 1994

作者： Lam, Louisa Suen, Ching Y. Centre for Pattern Recognition and Machine Intelligence Concordia University Montréal Canada

ISBN: (纸本)0818662700

Recently, it has been demonstrated that combining the decisions of several classifiers can lead to improved recognition results. The combination can be implemented using a variety of strategies, among which majority vote is by far the simplest, yet it has been found to be just as effective as more complicated schemes. However, all the results reported thus far on combinations of classifiers have been experimental in nature. The intention of this research is to analyze the foundations of the majority vote method in order to gain a deeper understanding and new results about its mode of operation. © 1994 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

关键词： pattern recognition systems

来源：评论

学校读者我要写书评

暂无评论

Noise tolerant script identification of printed oriental and English documents using a downgraded pixel density feature

Noise tolerant script identification of printed oriental and...

引用

2010 20th International Conference on pattern recognition, ICPR 2010

作者： Wang, Ning Lam, Louisa Suen, Ching Y. Centre for Pattern Recognition and Machine Intelligence Concordia University Montreal QC Canada

ISBN: (纸本)9780769541099

Document Script Identification (DSI) is a very useful application in document processing. This paper presents a method for this application that uses a new noise tolerant feature, the Downgraded Pixel Density feature. Compared to other features widely used in existing DSI solutions, this new feature is much more robust to variations in slant, font and style of printed documents. Experimental results show that the method achieves promising identification performances. © 2010 IEEE.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Face recognition using interest points and ensemble of classifiers 4

Face recognition using interest points and ensemble of class...

引用

4th IEEE International Conference on Recent Advances in Information Technology, RAIT 2018

作者： Vinay, A. Sampat, Pratik Rajesh Belavadi, Sagar V. Pratik, R. Rao, B. S. Nikitha Ragesh, Rahul Murthy, K. N. Balasubramanya Natarajan, S. Centre for Pattern Recognition and Machine Intelligence PES University Bangalore Karnataka India

ISBN: (纸本)9781538630396

In human beings, it is the responsibility of the temporal lobe of the brain for recognition of faces. Certain features of the face trigger the neurons of the temporal lobe which are then stored. These eventually lead to the identification of the face. In machine learning system, a huge dataset of images is used as a reference by the face recognition algorithms. The most prominent features are extracted from what is referred to as landmarks on the face and are analyzed based on their relative position amongst other features. The main challenge in computer vision lies in the identification and extraction of these features which in turn assist in face recognition owing to the exponentially large degree of variability that can be seen in appearances. Speed and accuracy are some of the most prominent challenges for the same. We explore ELM critically and study its performance. Owing to its extremely fast learning, it finds potential in real-time applications. The goal of this paper is to evaluate the performance of the extreme learning machine algorithm for facial recognition when used as a standalone model as well as when it is used with an ensemble learner. We then conclude with a voting classifier which is responsible for the final predictions made. Various combinations of ensembles are explored to come up with the best performing model. © 2018 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Face recognition Based on Adaptive Sparse Coefficient

引用

Journal of Donghua University(English Edition) 2016年第1期33卷 98-104页

作者：李洪均徐子能 SUEN Ching-yee School of Electronic Information Engineering Nantong University Centre for Pattern Recognition and Machine Intelligence Concordia University

Recently,robust sparse coding achieves high recognition rates on face recognition( FR),even when dealing with occluded images. However,robust sparse coding is that the coefficients are guaranteed global sparse when solving the sparse coefficients. In this paper,the coefficient vector is divided into multiple regions. Then,the elements in the object region are enabled to approximate global maximum by adding two constraint conditions( the maximal element of coefficient vector is in the object region; the sum of elements in the object region is the maximum value among all regions),which makes the distribution of sparse coefficient adapt to different classes of testing images. The efficacy of the proposed approach is verified on publicly available databases( i. e.,AR and Extended Yale B).Furthermore, the proposed method still can achieve a good performance when the training samples are limited.

关键词： face recognition(FR) adaptive sparse coefficient occlusion

来源：评论

学校读者我要写书评

暂无评论

Residual and Ensemble Learning on Locally Derived Features for Face recognition 1

引用

5th International Symposium on Intelligent Systems Technologies and Applications, ISTA 2019

作者： Vinay, A. Gupta, Abhijay Bharadwaj, Aprameya Srinivasan, Arvind Balasubramanya Murthy, K.N. Natarajan, S. Centre for Pattern Recognition and Machine Intelligence PES University BengaluruKA560085 India

ISBN: (数字)9789811539145

ISBN: (纸本)9789811539138

In this paper, an efficient approach toward face recognition is presented. The proposed model is invariant to pose, expression, scale, illumination, and translation with the application of different techniques and implementation of their algorithms. The images are first preprocessed using different techniques like CS-LBP, XCS-LBP, CS-LDMP, etc. to remove noise and make it illumination invariant. In one implementation, these are fed as input to different descriptors. The keypoints detected are invariant to lighting condition. A wide variety of descriptors are applied to the detected keypoints in order to compare the quality of features computed. These descriptors are quantized to a single vector, representing the most salient features of the facial image. The vectors are then fed into the stacking classifier. In the other method, these preprocessed images are directly fed as input into the residual network. A comparison of the results by these two pipelines is studied on two benchmark face datasets, namely FACES94 and Grimace. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

A hybrid multiple classifier system of unconstrained handwritten numeral recognition

引用

pattern recognition and Image Analysis 2007年第4期17卷 608-611页

作者： He, C.L. Suen, C.Y. Centre for Pattern Recognition and Machine Intelligence Concordia University Montreal Que. H3G IM8 Canada

o raise the reliability, a hybrid multiple classifier system is proposed by integrating the cooperation and combination of three classifiers: SVM [1], MQDF [3], and leNet5 [2]. In combination, we apply the total probability theorem to the classifiers at the rank level. Meanwhile, differential measurement and probability measurement are defined for the rejection option on different types of classifiers. Considerable improvement has been observed, and the final recognition rate of this system ranges from 95.54 to 99.11% with a reliability of 99.54 to 99.11%. © 2007 Pleiades Publishing, Ltd.

关键词： Image recognition

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：