Presentation 2008-05-23
How fast is an FPGA in image processing ?
Takashi SAEGUSA, Tsutomu MARUYAMA, Yoshiki YAMAGUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In image processing, FPGAs have shown very high performance in spite of their slow operational frequency. The main reasons for the high performance are (1) high parallelism in the applications in image processing, (2) the high ratio of 8 bit operations, and (3) parallel accesses to a large number of internal memory banks. In the recent micro processors, it becomes possible to execute SIMD instructions on 128 bit data in one clock cycle. Furthermore, these processors support multi-cores, and large cache memory which can hold all image data for each core. In this paper, we compare the performance of FPGAs with those processors in image processing using three applications; two-dimensional filters, stereo-vision and k-means clustering, and make it clear how an FPGA is fast in image processing, and how many hardware resources are required to achieve the performance.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) image processing / FPGA / SIMD processing
Paper # RECONF2008-15
Date of Issue

Conference Information
Committee RECONF
Conference Date 2008/5/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Reconfigurable Systems (RECONF)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) How fast is an FPGA in image processing ?
Sub Title (in English)
Keyword(1) image processing
Keyword(2) FPGA
Keyword(3) SIMD processing
1st Author's Name Takashi SAEGUSA
1st Author's Affiliation Systems and Information Engineering, University of Tsukuba()
2nd Author's Name Tsutomu MARUYAMA
2nd Author's Affiliation Systems and Information Engineering, University of Tsukuba
3rd Author's Name Yoshiki YAMAGUCHI
3rd Author's Affiliation Systems and Information Engineering, University of Tsukuba
Date 2008-05-23
Paper # RECONF2008-15
Volume (vol) vol.108
Number (no) 48
Page pp.pp.-
#Pages 6
Date of Issue