Performance Evaluation of Super High Definition Image Processing on a Parallel DSP System (Special Section of Papers Selected from the 7th Digital Signal Processing Symymposium
スポンサーリンク
概要
- 論文の詳細を見る
In this paper, we evaluate the sustained performance of the prototype SHD (Super High Definition) image processing system NOVI-II HiPIPE, and discuss the requirements of a real-time SHD image processing system. NOVI-II HiPIPE is a parallel DSP system with 128 PEs (Processing Elements), each containing one vector processor, and its peak performance is 15 GFLOPS. The measured performance of this system is at least 100 times higher than that of the Cray-2 (single CPU), but is still insufficient for real-time SHD image coding. When coding SHD moving images at 60 frames per second with the JPEG algorithm, the performance must be at least ten times faster than is now possible with NOVI-II HiPIPE. To extract higher performance from a parallel processing system, the system architecture must be suitable for the implemented process. The advantages of NOVI-II HiPIPE are its mesh network and high performance pipelined vector processor (VP), one of which is installed on each PE. When most basic SHD image coding techniques are implemented on NOVI-II HiPIPE, intercommunication occurs only between directly connected PEs, and its cost is very low. Each VP can efficiently execute vector calculations, which occur frequently in image processing, and they increase the performance of NOVI-II HiPIPE by a factor of from 20 to 100. In order to further improve the performance, the speed of memory access and bit operations must be increased. The next generation SHD image processing system must be built around the VP, an independent function block which controls memory access, and another block which executes bit operations. To support the input and output of SHD moving images and the inter-frame coding algorithms, the mesh network should be expanded into a 3D-cube.
- 社団法人電子情報通信学会の論文
- 1993-08-25
著者
-
Fujii Tatsuya
Ntt Network Innovation Laboratories
-
SAWABE Tomoko
NTT Network Innovation Laboratories
-
Sawabe Tomoko
NTT Transmission Systems Laboratories
-
Fujii Tatsuya
NTT Transmission Systems Laboratories
-
Fujii Tetsurou
NTT Transmission Systems Laboratories
-
Ono Sadayasu
NTT Transmission Systems Laboratories
関連論文
- Bit Allocation Method for Independently Encoded JPEG2000 Colour Components
- Bit Allocation Method for Independently Encoded JPEG2000 Colour Components
- Bit Allocation Method for Independently Encoded JPEG2000 Colour Components
- Multiresolution Lossless Video Coding Using Inter/Intra Frame Adaptive Prediction(Special Section on Digital Signal Processing)
- Performance Evaluation of Super High Definition Image Processing on a Parallel DSP System (Special Section of Papers Selected from the 7th Digital Signal Processing Symymposium
- Parallel Processing of Distributed Video Coding to Reduce Decoding Time
- Layered Low-Density Generator Matrix Codes for Super High Definition Scalable Video Coding System
- A Unified Coding Algorithm of Lossless and Near-Lossless Color Image Compression(Special Section on Intelligent Signal and Image Processing)
- Pel Adaptive Predictive Coding Based on Image Segmentation for Lossless Compression
- Lossless and Near-Lossless Color Image Coding Using Edage Adaptive Quantization
- Super High Definition Image Communications-A Platform for Media Integration (Special Issue on Next Generation Visual Telecommunication and Broadcasting)
- 超高精細動画フィルム入力システム(次世代画像技術)
- A Study on Non-octave Scalable Image Coding and Its Performance Evaluation Using Digital Cinema Test Material(Image)
- Extended Multiresolution Lossless Video Coding Using In-Band Spatio-Temporal Prediction (Image/Vision Processing, Multidimensional Signal Processing and Its Application)
- Layered Multicast Encryption of Motion JPEG2000 Code Streams for Flexible Access Control
- Layered Multicast Encryption of Motion JPEG2000 Code Streams for Flexible Access Control