首页 | 本学科首页   官方微博 | 高级检索  
     检索      

快速二维小波变换VLSI结构
引用本文:田昕,马丽,田金文.快速二维小波变换VLSI结构[J].华中科技大学学报(自然科学版),2009(9).
作者姓名:田昕  马丽  田金文
作者单位:华中科技大学多谱信息处理技术国防科技重点实验室;
基金项目:国家自然科学基金资助项目(60572048); 湖北省自然科学基金资助项目(2006ABA088)
摘    要:提出了一种快速有效的二维小波变换超大规模集成电路(VLSI).该结构是一种4输入/4输出直接型结构,行列滤波同时运行,包含4个行滤波单元和1个列滤波单元,行滤波单元通过使用折叠结构减少硬件资源,列滤波单元在每个时钟周期,可同时处理4路行滤波的结果.整个结构无需额外的缓存,数据处理无需停顿.将之与其他类似结构进行了比较,结果表明本结构系统响应快、输出速率高,适应于高速运算等应用场合.

关 键 词:小波变换  并行运算  超大规模集成电路  图像压缩  提升方法  流水线技术  

High-speed VLSI architecture for two-dimensional discrete wavelet transform
Tian Xin Ma Li Tian Jinwen.High-speed VLSI architecture for two-dimensional discrete wavelet transform[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,2009(9).
Authors:Tian Xin Ma Li Tian Jinwen
Institution:Tian Xin Ma Li Tian Jinwen (State Key Laboratory for Multi-spectral Information Processing Technology,Huazhong University of Science and Technology,Wuhan 430074,China)
Abstract:A fast and efficient very large scale integration(VLSI) architecture for two-dimensional discrete wavelet transform was proposed,which is a 4-input/4-output architecture.The row and column filter operations could be computed in parallel way.It included four row filters and one column filter.The row filter was implemented by folding technology to reduce the hardware cost.The column was a 4-input/4-output architecture.Four row filter results could be processed in each clock cycle.No extra storage was needed.T...
Keywords:wavelet transform  parallel computing  very large scale integration(VLSI)  image compression  lifting schemes  pipeline technique  
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号