DocumentCode :
2226070
Title :
Low entropy block matching algorithm for motion estimation
Author :
Qiu, Xiaohai ; Zhang, Wujian ; Chen, Hongyi ; Zhou, Runde
Author_Institution :
Inst. of Microelectron., Tsinghua Univ., Beijing, China
fYear :
2001
fDate :
2001
Firstpage :
405
Lastpage :
408
Abstract :
A new block matching algorithm is proposed in this paper. Walsh-hadamard Transform (WHT) is employed to search the matched block with low entropy in motion estimation and improves the encoding efficiency. Due to the properties of WHT, computation complexity and hardware cost of ME can be greatly reduced. Compared with the conventional full search, the proposed algorithm can result in high image quality (degradation < 0.1 dB for PSNR) with lower operation load (reduced to 1/50)
Keywords :
Hadamard transforms; Walsh functions; computational complexity; entropy; image coding; image matching; motion estimation; Walsh-Hadamard transform; block matching algorithm; computation complexity; encoding efficiency; entropy; hardware cost; image quality; motion estimation; operation load; Costs; Degradation; Discrete cosine transforms; Entropy; Hardware; Image coding; Image quality; Microelectronics; Motion estimation; PSNR;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
ASIC, 2001. Proceedings. 4th International Conference on
Conference_Location :
Shanghai
Print_ISBN :
0-7803-6677-8
Type :
conf
DOI :
10.1109/ICASIC.2001.982586
Filename :
982586
Link To Document :
بازگشت