码书

一种快速高效的VQ初始码书设计方法

A new rapid and high quality VQ initial codebook generating method

基于分频算法的VQ码书生成方法

An Algorithm to Generate VQ Codebook Based on Frequency Dividing

使用遗传算法（GA）与模糊c均值聚类（FCM）算法相结合的方法来设计码书，有效地克服了FCM算法容易陷入局部最优且对初始值敏感的缺点。

Codebook of vector quantization is designed through fuzzy c-mean clustering and genetic algorithm . This algorithm overcomes the shortcomings of FCM effectively .

基于MFCC和VQ码书的说话人识别系统研究

Research on a Speaker Recognition System using MFCC Features and Vector Quantization

即首先对图像进行小波分解，然后对高频子图像用改进后的LBG算法形成码书，进行VQ编码。

Firstly , the image is decomposed by discrete wavelet transform .

基于SOFM神经网络的码书设计及其在图像传输中的应用

Codebook Design Based on SOFM Neural Network and Its Application on Image Transmission

本文提出了基于改进禁止搜索（TS）算法的矢量量化（VQ）码书设计方法。

Codebook design algorithms based on tabu search ( TS ) approach are presented for vector quantization ( VQ ) .

LBG算法使用等价范数产生相同码书

LBG Algorithm with Equivalence Norms Generates Same Codebook

自适应码书搜索替代长时预测，是大多CELP编码方案的关键。

Adaptive code-book search plays an important role in most CELP by replacing long-term prediction .

本文提出了一种新型的基于对LSF矢量空间进行拉伸变化的混合进化码书优化算法。

A new hybrid evolutionary LSF codebook optimization algorithm is proposed based on the vector space stretching .

LBG算法是经典的码书设计算法，是各类码书设计算法的基础。

LBG algorithm is the classical algorithm for codebook design and is foundation of other algorithms .

本文提出了一种采用均方误差（MSE）测度的矢量量化码书搜索的快速算法。

A fast codebook search method based on the mean squared error ( MSE ) distortion for vector quantization is presented .

为了有效消除复杂动态背景对运动物体检测的影响，提出一种新的基于HSV颜色空间的码书模型。

A new codebook model was proposed based on HSV color space to eliminate the effect of complex dynamic background in the moving object detection .

本文对此算法加以改进，并结合改进的基于PCC的初始码书生成算法，使得整体算法的性能得到了很大的提高。

In this paper , we make it improved , and improve the performance of entire algorithm combined with improved PCC algorithm greatly .

详述了传统码书训练算法LBG的特性以及训练码书的失真测度。

The characteristic of the traditional codebook training algorithm LBG and distortion measurement are also dwelt on .

LD-CELP语音编码中一种新的码书设计方法

A New Codebook Design Method in LD-CELP Speech Coding Algorithm

用增益精确值和归一化波形码书改进G.728

Improvement on G.728 by normalized shape codebook with exact gain

将遗传算法与LBG算法相结合，得到了一种矢量量化码书设计算法。

This paper proposes a codebook design method based on genetic algorithm ( GA ) and the LBG algorithm .

证明了著名的LBG算法使用等价范数，只要初始码书相同，则产生相同码书。

This paper proves that famous LBG algorithm with equivalence norm generates same codebook if the initial codebooks are same .

论文的具体研究成果如下：1、提出了一种基于类Haar特征的分层码书背景模型。

Main achievements of this thesis are as follows : 1 . We proposed a method called Hierarchical codebook background model based on haar-like features .

G.728中增益码书的研究

A Study of Gain Codebook in G.728

用IRIS数据集对算法进行了测试，并应用所给算法进行了用于图像压缩的量化码书设计。

The algorithm is tested and evaluated using the IRIS data set . The efficiency of the algorithm is also illustrated by its use in codebook design required for image compression based on vector quantization .

试验证明了此方法的有效性，解决了LBG算法局部最优的局限，获取更接近全局最优的码书。

The experimental results prove that the algorithm has better improved the expected distortion by overcoming the local optimality of LBG algorithm .

由于在不同的清/浊音判决模式下，LSF参数的统计特性不同，因此采用不同的码书对LSF参数矢量进行量化，提高了量化的质量。

Due to the different statistic distributions in different modes , the LSF vectors are quantized by different codebooks to improve the quality of quantization .

研究结合向量量化的DCT变换压缩编码，首先基于向量理论，研究向量量化（VQ）编码的搜索方法，并提出了分类快速搜索码书结构及其VQ编码方法；

First , based on vector theory the fast search method for vector quantization coding is discussed . Also a new codebook structure of fast classification search for codeword is proposed .

网格编码量化（trelliscodedquantizer，TCQ）是一种利用卷积编码，信号空间扩展信号间的欧氏距离及码书划分的量化方法。

Trellis coded quantizer ( TCQ ) is a quantization method which make use of convolutional encoding , signal space expanding to increase Euclidian distance and codebook partition .

由于一个大小为N的矢量量化码书中的码字可以确定N（N-1）/2条直线，因此这种方法可用较小的码书获得很高的编码精度。

Since a N sized vector quantization ( VQ ) codebook can determine N ( N-1 ) / 2 different straight lines , the new method can encode vectors with very high coding accuracy based on a rather small codebook .

这两种算法充分利用了红外图像小波分解后的系数分布特点来构造矢量，并相应结合传统的LBG算法和改进的LBG算法生成了各自具有代表性的码书。

After the analysis of infrared image wavelet coefficients , a vector quantization algorithm based on wavelet transform and the advanced one are proposed .

指出LBG算法的缺陷，并针对LBG[1]算法对初始码书敏感这一缺陷进行分析。

With three different initial codebook design methods , this paper verified the conclusion that the LBG algorithm is sensitive to the initial codebook .

试验表明，RLS算法在语音参数聚类中表现出了很好的性能，得到的码书质量优于经典的LBG算法，从而为设计全局准最优码书提供了一种新思路。

In addition , it is demonstrated to be more effective in the clustering for speech parameters and be obtained better codebook quality in comparison with LBG algorithm .