Fast algorithm for VQ codebook design - Communications, Speech and Vision, IEE Proceedings I

نویسنده

  • Sin-Horng Chen
چکیده

The paper presents a fast codebook training algorithm for vector quantisation. It uses an elimination rule, based on triangular inequality criteria, as well as the partial distortion elimination method, to relieve the computational burden of a conventional codebook training algorithm, including a binary codeword splitting algorithm for the initial codebook and the LBG recursive algorithm. Over 95% savings in both multiplication and addition operations were achieved in the simulation of a VQ codebook training of a ‘Lena’ image using 16-dimensional vectors.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Use of multiple vector quantisation for semicontinuous-HMM speech recognition - Vision, Image and Signal Processing, IEE Proceedings-

Although the continuous hidden Markov model (CHMM) technique seems to be the most flexible and complete tool for speech modelling, it is not always used for the implementation of speech recognition systems because of several problems related to training and computational complexity. Thus, other simpler types of HMMs, such as discrete (DHMM) or semicontinuous (SCHMM) models, are commonly utilise...

متن کامل

Fast VQ codebook search algorithm for grayscale image coding

In this paper, a fast codebook search algorithm that is equivalent to the full search algorithm for image vector quantization is proposed. The image encoding procedure of VQ needs to find the closest codewords in the codebook for image blocks. A great deal of computational cost is consumed in the image encoding procedure. The proposed algorithm employs two test conditions to speedup the image e...

متن کامل

A continuous VQ clustering algorithm for realtime speech recognition

This paper presents a continuous VQ clustering (CVQC) algorithm for realtime speech recognition, which incorporates the temporal information of speech into both training and recognition processes. In comparison with the conventional DTW and VQ methods, this new algorithm delivers faster training and recognition speed and smaller codebook size while still retains merits of both. Realtime impleme...

متن کامل

Variable dimension VQ encoding and codebook design

A variable dimension vector quantizer (VDVQ) has codewords of unequal dimensions. Here, a trellis-based sequential optimal VDVQ encoding algorithm is proposed. Also, a VDVQ codebook design algorithm based on splitting a node with equal or reduced dimensions is proposed that does not require any codebook parameter to be prespecified unlike known schemes. The VDVQ system is shown to outperform a ...

متن کامل

Image Compression with Efficient Code Book Initialization Using Lbg Algorithm Image Compression with Efficient Codebook Initialization Using Lbg Algorithm

Vector quantization (VQ) has received a great attention in the field of multimedia data compression since last few decades because it has simple decoding structure and can provide high compression ratio. In general, algorithms of VQ codebook generation focus on solving two kinds of problem: (i) to determine the quantization regions and the code words that minimize the distortion error. (ii) to ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2004