A fast encoding algorithm for vector quantization based on weighted variance inequality and Hadamard transform

Vector quantization is one of high performance and popular methods for data compression. But it is extremely time consuming during the encoding process. In this paper, a fast encoding algorithm for vector quantization is proposed to save the computation time. This algorithm uses two characteristics...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2011 3rd International Conference on Computer Research and Development Ročník 4; s. 324 - 327
Hlavní autoři: Linbo Xie, Bang Huang
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.03.2011
Témata:
Algorithm design and analysis
Channel coding
codeword search
High temperature superconductors
HTS
Image coding
Signal processing algorithms
variance
Vector quantization
ISBN:1612848397, 9781612848396
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Vector quantization is one of high performance and popular methods for data compression. But it is extremely time consuming during the encoding process. In this paper, a fast encoding algorithm for vector quantization is proposed to save the computation time. This algorithm uses two characteristics of a vector, Hadamard transform (HT) and variance. The methods using one of these features was already proposed, they handles these features separately. Here, the proposed algorithm put forward a new inequality which utilizes these features simultaneously to rejects more codewords which are impossible to be the nearest codeword in the distortion computations stage. This method produces the same output as conventional full search algorithm. The simulation results show that the effectiveness of the proposed algorithm is outstanding.
ISBN:1612848397
9781612848396
DOI:10.1109/ICCRD.2011.5763885