搜索资源列表
HSADPCM
- 利用ADPCM技术压缩心音信号,函数的预测方法和量化码书基于IMA标准,对于输出误差还可以进一步编码压缩-Heart sound signals using the ADPCM compression technology, function prediction methods and quantization codebook based IMA standard for encoding the output error can
person-detection
- 采用混合高斯模型和codebook分别进行背景建模,实现行人检测-person detection, GMM and codebook
vector-quantization
- 用MATLAB或C语言实现码本训练和矢量量化算法,并给出量化结果-Using MATLAB or C language training and codebook vector quantization algorithm, and gives quantitative results
WAVELET-VQ
- 基于小波变换的VQ语音压缩编解码,首先对信号进行小波变换,再用VQLBG算法训练出码本,在信道两端模拟发送和接受,该源码完整,包含训练和测试语音-Wavelet-based VQ voice compression codec, first signal wavelet transform, and then VQLBG algorithm trained codebook simulations across the channel
c8_31pf
- Searches a 31 bit algebraic codebook containing : 8 pulses in a fr a me of 40 samples.: in the same manner as GSM-EFR.
CachedHistory
- Made some changes to the comments to match the comments from other modules.This file contains functions that reset and perform codebook gain calculations.
c2_9pf
- This file contains the functions that search a 9 bit algebraic codebook containing 2 pulses in a fr a me of 40 samples.
cbsearch
- Inovative codebook search (find index and gain).
Vector-quantization
- 采用自组织映射算法(SOM)设计码书,达到全局最优,同时采用组合码字和图像块动态划分等优化算法,最终实现的码书性能比LBG算法有很大提高。这是中科大视频技术的final project,附有最终版的report和presention。-Using self-organizing map algorithm (SOM) designed codebook, the global optimum, while using a combina
Embedded-Freebsd-Cookbook
- pdf book of Embedded Freebsd codebook, esay to read and folow the program flow-Embedded Freebsd codebook
VQ
- LBG算法设计矢量量化码书的函数,需要给出训练序列x,矢量维数K和码书长度CN,返回根据训练序列给出的最小误差码书。-LBG algorithm for vector quantization codebook design function , you need to give training sequence x, vector dimension K and codebook length CN, return accordin
codebook
- cedebook背景建模方法的背景差分方法检测运动目标!-cedebook background modeling method for background subtraction method for detecting moving targets!
codebook_learningcode
- codebook高级背景建模的例程,很好的学习资料-codebook high background modeling routines, good learning materials
VQ
- 使用Splitting算法和GLA算法实现对图像的矢量量化及重新编码和重建。-Vector Quantization:To generate a 16-codeword codebook using Splitting Algorithm and GLA,then use this codebook to code the training image and reconstuct it.
speaker-recognition
- mfcc用于求取训练样本的特征值,其需要melbankm和enfr a me两个程序。 enfr a me用于将语音分帧,melbankm用于产生mel三角滤波器组,vqlbg用于获取说话人码书,disteu程序用于计算两个矩阵间的距离,-mfcc strike eigenvalues for the training sample, which requires melbankm and enfr a
shengwenshibie
- 声纹识别的实现,里面包含一些声纹的数据,用的的12Mel频谱倒谱系数,然后应用的是模糊K均值举了你算法构成的码本,识别在小样本情况下非常高-Voiceprint identification of implementation, which contains some of the voiceprint data, with the 12Mel spectrum Cepstral, and then apply the fuzzy K-
StandardCB
- 标准的codebook源码,,根据codebook论文编写, 网络上大多是简化版 给出一个标准版供各位研究学习-Standard codebook source, according to the written codebook papers, mostly on the network given a simplified version of the standard version for you research st
Codebook_model
- Codebook model 视频抠像 xp sp3 + vs2005 + OpenCV 2.3.1-For a more detailed explanation of a “codebook” model refer to [reference: Gary Bradski and Adrian Kaehler: Learning Opencv, September 2008: First Edition. p. 278].
VQSOM
- 利用SOM神经网络设计矢量量化的码书,结果优于传统的LBG方法,失真度较小。-SOM neural network design use vector quantization codebook, the result is better than traditional methods LBG, distortion is small.
BackgroundCodebookConstruction
- 背景分离码本建模法,在静态背景下检测运动物体,并提供了一段监控视频以供测试-Background Modeling and Subtraction by Codebook Construction. Supply a video for testing.