一种基于互信息的模糊粗糙分类特征基因快速选取方法

目前,支持向量机技术(SVM)在遥感信息获取中普遍受到参数选择不准确和小样本问题的制约。针对这些问题, 提出一种新的半监督集成SVM(EPS3VM)分类模型。模型一方面利用自适应变异粒子群优化算法对SVM参数寻优以提高基分类器精度(PSVM);另一方面采用自训练算法(Self-training),充分利用大量廉价的未标记样本产生性能差异的半监督分类器个体(PS3VM),其中,在未标记样本标注过程中,引入模糊聚类算法(Gustafson-kessel)来控制错误类别的输入,最后对个体分类器采用加权集成策略,以进一步提高分类模型的泛化能力。为了测试其性能,应用该模型进行多光谱遥感影像的土地覆盖分类实验,并与PSVM、PS3VM进行对比,分类精度从PSVM的88.48%提高到96.88%,Kappa系数由0.8546提高到0.9606。结果表明,EPS3VM在克服传统SVM参数选择不准确的同时,有效地应对了小样本问题,分类性能更优。     

支持向量机在低信噪比语音识别中的应用

采用改进的MFCC语音特征参数(Mel频率离散小波倒谱系数),使用支持向量机作为分类算法,构建了低信噪比环境下的孤立词非特定人语音识别系统,取得了较高的识别率。将实验结果与基于RBF神经网络的识别结果进行比较,结果表明在低信噪比时,SVM的识别率比使用RBF神经网络有较大提高,具有非常好的鲁棒性。     

Fall detection using single-tree complex wavelet transform

The goal of Ambient Assisted Living (AAL) research is to improve the quality of life of the elderly and handicapped people and help them maintain an independent lifestyle with the use of sensors, signal processing and telecommunications infrastructure. Unusual human activity detection such as fall detection has important applications. In this paper, a fall detection algorithm for a low cost AAL system using vibration and passive infrared (PIR) sensors is proposed. The single-tree complex wavelet transform (ST-C$\mathbb{C}$WT) is used for feature extraction from vibration sensor signal. The proposed feature extraction scheme is compared to discrete Fourier transform and mel-frequency cepstrum coefficients based feature extraction methods. Vibration signal features are classified into “fall” and “ordinary activity” classes using Euclidean distance, Mahalanobis distance, and support vector machine (SVM) classifiers, and they are compared to each other. The PIR sensor is used for the detection of a moving person in a region of interest. The proposed system works in real-time on a standard personal computer.     

基于Gabor小波与支持向量机对储粮害虫分类识别

将Gabor小波与支持向量机有机结合起来,对储粮害虫进行分类识别。首先对原始图像进行不同尺度和不同方向的Gabor滤波,获得原始图像特征,然后将遗传算法与支持向量机相结合来自动优选支持向量机模型参数,减少了以往应用中需反复实验来确定其参数的人工工作量。实验结果表明该方法识别率高,识别速度快,容错性好,而且能够正确识别有噪声的储粮害虫图像。为储粮害虫的快速鉴定和分类研究提供了可靠和科学的信息。     

Type-2 fuzzy logic-based classifier fusion for support vector machines

As a machine-learning tool, support vector machines (SVMs) have been gaining popularity due to their promising performance. However, the generalization abilities of SVMs often rely on whether the selected kernel functions are suitable for real classification data. To lessen the sensitivity of different kernels in SVMs classification and improve SVMs generalization ability, this paper proposes a fuzzy fusion model to combine multiple SVMs classifiers. To better handle uncertainties existing in real classification data and in the membership functions (MFs) in the traditional type-1 fuzzy logic system (FLS), we apply interval type-2 fuzzy sets to construct a type-2 SVMs fusion FLS. This type-2 fusion architecture takes considerations of the classification results from individual SVMs classifiers and generates the combined classification decision as the output. Besides the distances of data examples to SVMs hyperplanes, the type-2 fuzzy SVMs fusion system also considers the accuracy information of individual SVMs. Our experiments show that the type-2 based SVM fusion classifiers outperform individual SVM classifiers in most cases. The experiments also show that the type-2 fuzzy logic-based SVMs fusion model is better than the type-1 based SVM fusion model in general.     

A novel excitation controller using support vector machines and approximate models

This paper proposes a novel excitation controller using support vector machines （SVM） and approximate models. The nonlinear control law is derived directly based on an input-output approximation method via Taylor expansion, which not only avoids complex control development and intensive computation, but also avoids online learning or adjustment. Only a general SVM modelling technique is involved in both model identification and controller implementation. The robustness of the stability is rigorously established using the Lyapunov method. Several simulations demonstrate the effectiveness of the proposed excitation controller.     

A theory of computer instructions

Just as there is a theory of groups, or rings, or fields, or topological spaces, so there is presented here a theory of computer instructions. These are functions from to , where is the set of states of a computer. Here is a set of functions from M to B, where M is the memory (the set of variables) and B is a set of values; or alternatively, is a cartesian product, over M as an index set, of the sets of values of all variables in M. For each instruction I there are defined the input region IR(I) and the output region OR(I); these are subsets of M. An instruction takes data from its input region and places data in its output region. Here OR(I) may be decomposed further into regions affected by subsets of M, which may be defined in either of two alternative ways. The theory presented here includes theorems concerning composition of instructions, decomposition of instructions, and the existence of instructions with specified regions. Many examples are given.     

基于虚拟机技术构建网络技术实验平台

针对网络技术专业建立实验平台进行了探讨,提出了利用虚拟机技术,搭建网络技术实验平台。描述了网络实验环境的流程。     

Design and implementation of the virtual machine constructing on register

The technology of virtual machines is widely applied in many fields, such as code transplanting, cross-platform computing, and hardware simulation. The main purpose is to simulate the environment of general hardware by means of software and realize the operation of byte codes in processor. Virtual machines are commonly used to distribute programs in an architecture-neutral format, which can easily be interpreted or compiled. A long-running question in the design of virtual machines is whether stack architecture or register architecture can be implemented more efficiently with an interpreter. After introducing the basic theory, a solution to designing and realizing virtual machine constructing on register processor proposed in this paper, which gives a preliminary study on simulating the process of instructions in Intel x86.