检索结果-内蒙古大学图书馆

IEEE SoutheastCon 2005

作者： Chin, C Barreto, A Zhai, J Li, C Florida Int Univ Dept Elect & Comp Engn Miami FL 33199 USA

ISBN: (纸本)0780388658

At present, a three-input Electromyography (EMG) system has been created to provide real-time, hands-free cursor control. The system uses the real-time spectral analysis of three EMG signals to produce the following five cursor actions: i) LEFT, ii) RIGHT, iii) UP, iv) DOWN, v) LEFT-CLICK. The three EMG signals are obtained from two surface electrodes placed on the left and right temples of the head and one electrode placed in the forehead region. The present system for translating EMG activity into cursor actions does not always discriminate between up and down EMG activity efficiently. To resolve this problem it was proposed that the three-electrode system be converted into a four-electrode system, using two electrodes in the forehead of the user, instead of one. This paper compares the effectiveness of the four-electrode system to that of the three-electrode system in classifying EMG activity into cursor actions through the use of Matlab simulations. It will be shown that the new four-electrode system produces significant improvements in classification performance.

关键词： keyboards mouse controllers (computers) electromyography signal classification spectral analysis handicapped aids human computer interaction classification algorithm electromyography-based computer cursor control system three-input electromyography system EMG real-time hands-free cursor control real-time spectral analysis EMG signals cursor actions cursor left motion cursor right motion cursor up motion cursor down motion cursor left-click surface electrodes forehead region temple electrodes EMG activity translation activity discrimination four-electrode system Matlab simulations classification performance

来源：评论

学校读者我要写书评

暂无评论

Two improvements in genetic programming for image classification

Two improvements in genetic programming for image classifica...

引用

IEEE Congress on Evolutionary Computation

作者： Li, Yamin Ma, Jinru Zhao, Qiuxia Agr Univ Hebei Artificial Intelligence Res Ctr Baoding 071001 Peoples R China Agr Univ Hebei Coll Foreign Language Baoding 071001 Peoples R China

ISBN: (纸本)9781424418220

A new classification algorithm for mufti-image classification in genetic programming (GP) is introduced, which is the centered dynamic class boundary determination with quick-decreasing power value of arithmetic progression. In the classifier learning process using GP for mufti-image classification, different sets of power values are tested to achieve a more suitable range of margin values for the improvement of the accuracy of the classifiers. In the second development, the program size is introduced into the fitness function to control the size of program growth during the evolutionary learning process. The approach is examined on a Chinese character image data set and a grass leaves data set, both of which have four or more classes. The experimental results show that while dealing with complicated problems of mufti-image classification, the new approach can he used for more accurate classification and work better than the previous algorithms of either static or dynamic class boundary determination. With the fitness function, the size of the programs in the population can be controlled effectively and shortened considerably during evolution. Thus, the readability of the programs could be seemingly improved.

关键词： Genetic programming (GP) Image classification classification algorithm Fitness function construction

来源：评论

学校读者我要写书评

暂无评论

Optimizing unsupervised classifications of remotely sensed imagery with a data-assisted labeling approach

引用

COMPUTERS & GEOSCIENCES 2008年第12期34卷 1877-1885页

作者： Lang, Ruili Shao, Guofan Pijanowski, Bryan C. Farnsworth, Richard L. Purdue Univ Dept Forestry & Nat Resources W Lafayette IN 47907 USA

The quality of remotely sensed land use and land cover (LULC) maps is affected by the accuracy of image data classifications. Various efforts have been made in advancing Supervised or unsupervised classification a methods to increase the repeatability and accuracy of LULC mapping. This study incorporates a data-assisted labeling approach (DALA) into the unsupervised classification of remotely sensed imagery. The DALA-unsupervised classification algorithm consists of three steps: (1) creation of N spectral-class maps using Iterative Self-Organizing Data Analysis Technique algorithm (ISODATA);(2) development of LULC maps with assistance of reference data;and (3) accuracy assessments of all the LULC maps using independent reference data and selection or one LULC map with the highest accuracy. classification experiments with a composite image of a Landsat Thematic Mapper (TM) image and tin Enhanced Thematic Mapper Plus (ETM +) image suggest that DALA was effective in making unsupervised classification process more objective, automatic, and accurate. A comparison between the DALA-unsupervised classifications and some conventional classifications suggests that the DALA-unsupervised classification algorithm yielded better classification accuracies compared to these conventional approaches. Such a simple, effective approach has not been systematically examined before but has great potential for many applications in the geosciences. (C) 2008 Elsevier Ltd. All rights reserved.

关键词： LULC mapping Automated classification Landsat classification accuracy classification algorithm

来源：评论

学校读者我要写书评

暂无评论

URBAN RAIL TRAFFIC TICKETING INCOME DISTRIBUTION METHOD BASED ON SECURITY POLICY

引用

38th International Conference on Computers and Industrial Engineering

作者： Wang, Xiaoyong Lv, Yongbo Fu, Pengbo Jia, Nan Beijing Jiaotong Univ Sch Traff & Transportat Beijing 100044 Peoples R China

ISBN: (纸本)9787121074370

The traditional rail traffic ticket income distribution model has a different accuracy and scope of application and it is of no practical significance in the applications. In this paper, the improved distribution model of urban traffic ticketing income are designed in accordance with the principle of protection, the principle of proportion and the principle of execution by phase. In addition, the corresponding algorithms to the model are also solved. The improved model could address the distribution of the rail traffic ticketing income more effectively.

关键词： Distribution Rail traffic classification algorithm Security policy

来源：评论

学校读者我要写书评

暂无评论

Neuro-computing vector classification SVM schemes to integrate the overmodulation region in neutral point clamped, (NPC) converters

引用

IEEE TRANSACTIONS ON POWER ELECTRONICS 2007年第3期22卷 995-1004页

作者： Saeedifard, Maryarn Bakhshai, Alireza Univ Toronto Dept Elect & Comp Engn Toronto ON M5S 3G4 Canada Queens Univ Dept Elect & Comp Engn Kingston ON K7L 3N6 Canada

Three-level neutral point clamped (NPC) voltage source converters have recently emerged as important alternatives to conventional two-level converter topologies in high-power medium-voltage energy conversion applications, particularly in high performance ac motor drive systems. An all-inclusive modulation strategy for NPC converters should have the capability of extending the operating range of the converter into the overmodulation region with a smooth and linear transition characteristic. An overmodulation switching strategy based on the space vector classification technique for three-level NPC converters is introduced in this paper. The proposed overmodulation modes, make possible continuous control of the output voltage up to the maximum possible with a smooth linear transition characteristic, and minimum distortions. A theoretical basis for the vector classification space vector modulation technique in the overmodulation region is presented, and the proposed overmodulation schemes are validated by analysis, simulation and experimentation on a 2-KVA three-level NPC converter laboratory prototype.

关键词： classification algorithm overmodulation space vector modulation (SVM) three-level converter

来源：评论

学校读者我要写书评

暂无评论

Fully online classification by regularization

引用

APPLIED AND COMPUTATIONAL HARMONIC ANALYSIS 2007年第2期23卷 198-214页

作者： Ye, Gui-Bo Zhou, Ding-Xuan Fudan Univ Sch Math Sci Shanghai 200433 Peoples R China City Univ Hong Kong Dept Math Kowloon Hong Kong Peoples R China

In this paper we consider fully online learning algorithms for classification generated from Tikhonov regularization schemes associated with general convex loss functions and reproducing kernel Hilbert spaces. For such a fully online algorithm, the regularization parameter in each learning step changes. This is the essential difference from the partially online algorithm which uses a fixed regularization parameter. We first present a novel approach to the drift error incurred by the change of the regularization parameter. Then we estimate the error of the learning process for the strong approximation in the reproducing kernel Hilbert space. Finally, learning rates are derived from decays of the regularization error. The convexity of the loss function plays an important role in our analysis. Concrete learning rates are given for the hinge loss and the support vector machine q-norm loss. (c) 2007 Elsevier Inc. All rights reserved.

关键词： classification algorithm online learning reproducing kernel Hilbert spaces regularization error analysis

来源：评论

学校读者我要写书评

暂无评论

A stochastic algorithm for feature selection in pattern recognition

引用

JOURNAL OF MACHINE LEARNING RESEARCH 2007年 8卷 509-547页

作者： Gadat, Sebastien Younes, Laurent ENS Cachan CMLA F-94235 Cachan France Johns Hopkins Univ Ctr Imaging Sci Baltimore MD 21218 USA

We introduce a new model addressing feature selection from a large dictionary of variables that can be computed from a signal or an image. Features are extracted according to an efficiency criterion, on the basis of specified classification or recognition tasks. This is done by estimating a probability distribution P on the complete dictionary, which distributes its mass over the more efficient, or informative, components. We implement a stochastic gradient descent algorithm, using the probability as a state variable and optimizing a multi-task goodness of fit criterion for classifiers based on variable randomly chosen according to P. We then generate classifiers from the optimal distribution of weights learned on the training set. The method is first tested on several pattern recognition problems including face detection, handwritten digit recognition, spam classification and micro-array analysis. We then compare our approach with other step-wise algorithms like random forests or recursive feature elimination.

关键词： stochastic learning algorithms Robbins-Monro application pattern recognition classification algorithm feature selection

来源：评论

学校读者我要写书评

暂无评论

Multi-kernel regularized classifiers

引用

JOURNAL OF COMPLEXITY 2007年第1期23卷 108-134页

作者： Wu, Qiang Ying, Yiming Zhou, Ding-Xuan City Univ Hong Kong Dept Math Kowloon Hong Kong Peoples R China

A family of classification algorithms generated from Tikhonov regularization schemes are considered. They involve mufti-kernel spaces and general convex loss functions. Our main purpose is to provide satisfactory estimates for the excess misclassification error of these mufti-kernel regularized classifiers when the loss functions achieve the zero value. The error analysis consists of two parts: regularization error and sample error. Allowing mufti-kernels in the algorithm improves the regularization error and approximation error, which is one advantage of the mufti-kernel setting. For a general loss function, we show how to bound the regularization error by the approximation in some weighted L-q spaces. For the sample error, we use a projection operator. The projection in connection with the decay of the regularization error enables us to improve convergence rates in the literature even for the one-kernel schemes and special loss functions: least-square loss and hinge loss for support vector machine soft margin classifiers. Existence of the optimization problem for the regularization scheme associated with multi-kernels is verified when the kernel functions are continuous with respect to the index set. Concrete examples, including Gaussian kernels with flexible variances and probability distributions with some noise conditions, are used to illustrate the general theory. (c) 2006 Elsevier Inc. All rights reserved.

关键词： classification algorithm multi-kernel regularization scheme convex loss function misclassification error regularization error and sample error

来源：评论

学校读者我要写书评

暂无评论

On the construction of imputation classes in surveys

引用

INTERNATIONAL STATISTICAL REVIEW 2007年第1期75卷 25-43页

作者： Haziza, David Beaumont, Jean-Francois Univ Montreal Dept Math & Stat Montreal PQ H3C 3J7 Canada STAT Canada Business Survey Methods Div Ottawa ON Canada

This paper explores the problem of the construction of imputation classes using the score method, sometimes called predictive mean stratification or response propensity stratification, depending on the context. This method was studied in Thomsen (1973), Little (1986) and Eltinge & Yansaneh (1997). We use a different framework to evaluate the properties of the resulting imputed estimator of a population mean. In our framework, we condition on the realized sample. This enables us to considerably simplify our theoretical developments in the frequent situation where the boundaries and the number of classes are sample-dependent. We find that the key factor for reducing the non-response bias is to form classes homogeneous with respect to the response probabilities and/or the conditional expectation of the variable of interest. In the latter case, the non-response/imputation variance is also reduced. Finally, we performed a simulation study to fully evaluate various versions of the score method and to compare them with a cross-classification method, which is frequently used in practice. The results showed the superiority of the score method in general.

关键词： classification algorithm cross-classification method imputation model non-response and imputation variance non-response model non-response bias score method

来源：评论

学校读者我要写书评

暂无评论

Automatic differentiation of melanoma and clark nevus skin lesions

Automatic differentiation of melanoma and clark nevus skin l...

引用

Medical Imaging 2007 Conference

作者： LeAnder, R. W. Kasture, A. Pandey, A. Umbaugh, S. E. So Illinois Univ Edwardsville Dept Elect & Comp Engn Edwardsville IL 62026 USA

ISBN: (纸本)9780819466327

Skin cancer is the most common form of cancer in the United States. Although melanoma accounts for just 11% of all types of skin cancer, it is responsible for most of the deaths, claiming more than 7910 lives annually. Melanoma is visually difficult for clinicians to differentiate from Clark nevus lesions which are benign. The application of pattern recognition techniques to these lesions may be useful as an educational tool for teaching physicians to differentiate lesions, as well as for contributing information about the essential optical characteristics that identify them. Purpose: This study sought to find the most effective features to extract from melanoma, melanoma in situ and Clark nevus lesions, and to find the most effective pattern-classification criteria and algorithms for differentiating those lesions, using the Computer Vision and Image Processing Tools (CVIPtools) software package. Methods: Due to changes in ambient lighting during the photographic process, color differences between images can occur. These differences were minimized by capturing dermoscopic images instead of photographic images. Differences in skin color between patients were minimized via image color normalization, by converting original color images to relative-color images. Relative-color images also helped minimize changes in color that occur due to changes in the photographic and digitization processes. Tumors in the relative-color images were segmented and morphologically filtered. Filtered, relative-color, tumor features were then extracted and various pattern-classification schemes were applied. Results: Experimentation resulted in four useful pattern classification methods, the best of which was an overall classification rate of 100% for melanoma and melanoma in situ (grouped) and 60% for Clark nevus. Conclusion: Melanoma and melanoma in situ have feature parameters and feature values that are similar enough to be considered one class of tumor that significantly differs from

关键词： skin cancer dermoscopic images preprocessing color normalization relative-color images quantitative image analysis tumor texture features tumor identification tumor differentiation classification algorithm

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：