The recognition of a person from his or her gait has been a recent focus in computer vision because of its unique advantages such as being non-invasive and human friendly. However, gait recognition is not as reliable ...
详细信息
The recognition of a person from his or her gait has been a recent focus in computer vision because of its unique advantages such as being non-invasive and human friendly. However, gait recognition is not as reliable an identifier as other biometrics. In this paper, we applied a hierarchical fair competition-based parallel genetic algorithm and a neural network ensemble to the gait recognition problem. A diverse set of potential neural networks are generated to increase the reliability of the gait recognition, not only the best ones. Furthermore, a set of component neural networks is selected to build a gait recognition system such that generalization errors are minimized and negative correlation is maximized. Experiments are carried out with the NLPR and SOTON gait databases and the effectiveness of the proposed method for gait recognition is demonstrated and compared to previous methods.
In this paper, we develop a design methodology for information granulation-basedgenetically optimized fuzzy inference system, which deals with the tuning method with a variant identification ratio for structural as w...
详细信息
In this paper, we develop a design methodology for information granulation-basedgenetically optimized fuzzy inference system, which deals with the tuning method with a variant identification ratio for structural as well as parametric optimization of the reasoning system. The tuning is carried out with the aid of the hierarchical fair competition-based parallel genetic algorithms and it employs the mechanism of information granulation. This version of the geneticalgorithm is a multi-population variant of parallelgeneticalgorithms, which is particularly suitable for handling multimodal problems of high-dimensionality. The granulation of information is realized with the aid of the C-Means clustering algorithm. The concept of information granulation is applied to the formation of the fuzzy inference system in order to realize its structural optimization. Here we divide the input space in order to construct the premise part of the fuzzy rules. Subsequently the consequence part of each fuzzy rule is organized based on the center points (prototypes) of data group obtained as a result of clustering. In particular, this concerns the fuzzy inference system-related parameters, i.e., the number of input variables to be used in the fuzzy inference system, a collection of a specific subset of input variables. the number of membership functions used for each input variable, and the polynomial type (order) occurring at the consequence part of fuzzy rules. Making use of a mechanism of simultaneous tuning for the parameters, we construct an optimized fuzzy inference system related to its structural as well as parametric optimization. A comparative analysis demonstrates that the proposed methodology leads to improved results when compared with some conventional methods exploited in fuzzy modeling. (C) 2008 Elsevier Inc. All rights reserved.
An ensemble of neural networks exhibits higher generalization performance compared to a single neural network. In this paper, a new design method for a neural network ensemble is proposed. The hierarchical pair compet...
详细信息
An ensemble of neural networks exhibits higher generalization performance compared to a single neural network. In this paper, a new design method for a neural network ensemble is proposed. The hierarchical pair competition-basedparallelgeneticalgorithm (HFC-PGA) is employed to train the neural networks forming the ensemble. The aim of the HFC-PGA is to achieve not only the best neural network, but also a diversity of potential neural networks. A set of component neural networks is selected to build an ensemble such that the generalization error is minimized and the negative correlation is maximized. Finally, some experiments are carried out using several data sets to illustrate and quantify the performance of the proposed method.
暂无评论