Implementation of real-time 2D-DCT with FPGA and DSP

Bao Jun Zhao*, Cai Cheng Shi, Li Bi, Jian Bo An, Er Ke Mao

*此作品的通讯作者

科研成果: 期刊稿件文章同行评审

11 引用 (Scopus)

摘要

Because of the FPGA's parallel pipelining processing features, this paper designed and implemented the real time GIF format image DCT using Exilinx Company' s 500000 gate grade chip XCV400E. Using ping-pong model, C × F × CT 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 input and is multiplied CT 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 × (l × 8) × (8 × 8) times matrix operations. 44 times (1 × 8) × (8 × 8) matrix operations results need storing as transform format (HT=(F × CT)T). 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 × CT), they are read out and processed as (HT × CT). The final results are output as transform format (GT=C × H). Therefore, the continuous real-time whole field pix DCT transform C × F × CT is finished. Function and timing simulation and the successful connection with TMS320C62X system verified the design and implement.

源语言英语
页(从-至)1317-1319
页数3
期刊Tien Tzu Hsueh Pao/Acta Electronica Sinica
31
9
出版状态已出版 - 9月 2003

指纹

探究 'Implementation of real-time 2D-DCT with FPGA and DSP' 的科研主题。它们共同构成独一无二的指纹。

引用此