Implementation of Real-Time 2D-DCT with FPGA and DSP

ZHAO Bao-jun; SHI Cai-cheng; BI li; AN Jian-bo; MAO Er-ke

您当前的位置：

首页 >

文章列表页 >

Implementation of Real-Time 2D-DCT with FPGA and DSP

更新时间：2025-07-16

- Implementation of Real-Time 2D-DCT with FPGA and DSP
- Acta Electronica Sinica Vol. 31, Issue 9, Pages: 1317-1319(2003)
- 作者机构：
  
  北京理工大学电子工程系,北京,100081
- 作者简介：
- 基金信息：
- DOI：
  CLC： TN957
- Published：2003
- 稿件说明：
移动端阅览
ZHAO Bao-jun, SHI Cai-cheng, BI li, et al. Implementation of Real-Time 2D-DCT with FPGA and DSP[J]. Acta Electronica Sinica, 2003, 31(9): 1317-1319.
DOI：

ZHAO Bao-jun, SHI Cai-cheng, BI li, et al. Implementation of Real-Time 2D-DCT with FPGA and DSP[J]. Acta Electronica Sinica, 2003, 31(9): 1317-1319. DOI：

摘要

利用FPGA的并行分布流水特点

选用exilinx公司的50万门级芯片XCV400E

设计并实现CIF格式(352×288象素)图像实时DCT变换.该设计采用乒乓模式

只需设计一个快速算法模块

(F×C

)就解决了C×F×C

的实现算法.当视频信号通过数字化后逐行输入FPGA

在行、场同步信号和采样时钟的控制下

每输入一组数据(8个)

就进行行向量与

的矩阵乘运算(

F×C

)

并将结果按转置方式保存

每输入一个数进行一次(1×8)×(1×8)矩阵运算

每行进行352×(1×8)×(8×8)次矩阵运算

其中44次(1×8)×(8×8)矩阵运算的结果需要按转置形式(

=(F×C

)

)存储;当输入下一组8行数据时

对该组数据进行与前述8行数据相同的矩阵运算

而对刚做完(

F×C

)运算的8行相应结果

则按正常顺序取出进行(

×C

)运算

将结果按转置形式(

=C×H

)输出.从而以实时流水的方式完成

C×F×C

运算.功能仿真、时序仿真和与TMS320C62X系统的成功对接验证了本设计及算法的正确性.

Abstract

Because of the FPGA's parallel pipelining processing features

this paper designed and implemented the real time CIF format image DCT using Exilinx Company's 500000 gate grade chip XCV400E.Using ping-pong model

C×F×C

is implemented only by designing one fast algorithm model (

F×C

).Digital video signal is input to FPGA line by line.Controlling by horizontal sync and vertical sync

every group data of 8 pixels as a vector is inp

ut and is multiplied

i.e.(

F×C

).The computed results are stored as transform format.Each pixel needs one (1×8)×(8×8) matrix operations.Each line needs 352×(1×8)×(8×8) times matrix operations.44times (1×8)×(8×8) matrix operations results need storing as transform format (

=(F×C

)

).When next 8 line data are input

they are processed with the same way as the above.For the last 8 line's first processed results (

F×C

)

they are read out and processed as (

×C

).The final results are output as transform format (

=C×H

).Therefore

the continuous real-time whole field pix DCT transform

C×F×C

is finished.Function and timing simulation and the successful connection with TMS320C62X system verified the design and implement.

关键词

Keywords

references

Views

1806

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research of Data-Parallel-Implementation Method of DCT Based on SIMD-PE-Array

An Algorithm of Multilevel Gradient Energy based Face Detection in DCT Compressed Domain

Block DCT Compression of Triangle Meshes

Image Retrieval Based on DCT Compressed-Domain

Still Image Compression Using Embedded DCT Approach

Related Author

ZHONG Sheng

SHEN Lan-sun

LI Xiao-hua

LI Xiao-guang

ZHANG Hong-bin

LIU Bo

HUANG Xiang-lin

SHEN Lan-sun

Related Institution

Xi'an Microelectronic Technique InstituteXi'anShaanxi 710075China

Xi'an Microelectronic Technique Institute

北京工业大学信号与信息处理研究室

Computer Institute,Beijing University of Technology

Signal and Information Processing Lab, Beijing Polytechnic University

⁰