ASIP on Multi-Processor Architectures for Deblocking Filters
碩士 === 逢甲大學 === 資訊工程所 === 100 === The deblocking filters are used to remove the discontinuous phenomenon in the H.264/AVC technology so that the image quality may be improved in a high compression rate. However, the deblocking filter occupies the one-third computation overhead of the H.264/AVC codec...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2012
|
Online Access: | http://ndltd.ncl.edu.tw/handle/12541925851465488436 |
id |
ndltd-TW-100FCU05392082 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-100FCU053920822015-10-13T21:27:32Z http://ndltd.ncl.edu.tw/handle/12541925851465488436 ASIP on Multi-Processor Architectures for Deblocking Filters 特殊應用指令集多處理器架構於去區塊濾波器的應用 Jun-xiong Wu 吳俊雄 碩士 逢甲大學 資訊工程所 100 The deblocking filters are used to remove the discontinuous phenomenon in the H.264/AVC technology so that the image quality may be improved in a high compression rate. However, the deblocking filter occupies the one-third computation overhead of the H.264/AVC codec. This thesis proposes application-specific instruction-set multi-processor architectures to improve the deblocking filter performance with slight design and hardware overheads. Firstly, we partition the input data of the deblocking algorithm, and then assign each partition to individual processor to get parallel computing benefits. The application-specific instructions are selected for each partition of codes to increase performance. none 王益文 2012 學位論文 ; thesis 48 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 逢甲大學 === 資訊工程所 === 100 === The deblocking filters are used to remove the discontinuous phenomenon in the H.264/AVC technology so that the image quality may be improved in a high compression rate. However, the deblocking filter occupies the one-third computation overhead of the H.264/AVC codec.
This thesis proposes application-specific instruction-set multi-processor architectures to improve the deblocking filter performance with slight design and hardware overheads. Firstly, we partition the input data of the deblocking algorithm, and then assign each partition to individual processor to get parallel computing benefits. The application-specific instructions are selected for each partition of codes to increase performance.
|
author2 |
none |
author_facet |
none Jun-xiong Wu 吳俊雄 |
author |
Jun-xiong Wu 吳俊雄 |
spellingShingle |
Jun-xiong Wu 吳俊雄 ASIP on Multi-Processor Architectures for Deblocking Filters |
author_sort |
Jun-xiong Wu |
title |
ASIP on Multi-Processor Architectures for Deblocking Filters |
title_short |
ASIP on Multi-Processor Architectures for Deblocking Filters |
title_full |
ASIP on Multi-Processor Architectures for Deblocking Filters |
title_fullStr |
ASIP on Multi-Processor Architectures for Deblocking Filters |
title_full_unstemmed |
ASIP on Multi-Processor Architectures for Deblocking Filters |
title_sort |
asip on multi-processor architectures for deblocking filters |
publishDate |
2012 |
url |
http://ndltd.ncl.edu.tw/handle/12541925851465488436 |
work_keys_str_mv |
AT junxiongwu asiponmultiprocessorarchitecturesfordeblockingfilters AT wújùnxióng asiponmultiprocessorarchitecturesfordeblockingfilters AT junxiongwu tèshūyīngyòngzhǐlìngjíduōchùlǐqìjiàgòuyúqùqūkuàilǜbōqìdeyīngyòng AT wújùnxióng tèshūyīngyòngzhǐlìngjíduōchùlǐqìjiàgòuyúqùqūkuàilǜbōqìdeyīngyòng |
_version_ |
1718063226259767296 |