High-Performance Designs for Linear Algebra Operations on Image Processing
DOI:
https://doi.org/10.20894/IJMSR.117.001.001.008Keywords:
Salt and Pepper, Adaptive, Pipeline, FIFO, Rank order, Non linearAbstract
Numerical linear algebra operations are key primitives in scientific computing. Performance optimizations of such operations have been extensively investigated. With the rapid advances in technology, hardware acceleration of linear algebra applications using field-programmable gate arrays (FPGAs) has become feasible. In this paper, we propose FPGA-based designs for several basic linear algebra operations, including dot product, matrix-vector multiplication, matrix multiplication, and matrix Factorization. By identifying the parameters for each operation, we analyze the trade-offs and propose a high-performance design. In the implementations of the design a convolution filter using dot product, this is especially useful in noise removal. The proposed designs are implemented on Xilinx FPGAs. Experimental results show that our designs scale with the available hardware resources. Also, the performance of our designs compares favorably with that of general-purpose processor-based designs.
Downloads
References
2. Olli Vainio, Yrjö Neuvo, Steven E. Butner, A Signal Processor for Median-Based Algorithms, IEEE Transactions on Acoustics, Speech, Processing VOL 37. NO. 9, September 1989.
3. V.V. Bapeswara Rao and K. Sankara Rao, A New Algorithm for Real-Time Median Filtering, IEEE Transactions on Acoustics, Speech, Processing VOL ASSP-34. NO. 6, December 1986.
4. M. O. Ahmad and D. Sundararajan, Parallel Implementation of a Median Filtering Algorithm, Int. Symp. on Signals and Systems, 1988.
5. Dobrowiecki Tadeusz, Medián Szűrők, Mérés és Automatika, 37. Évf., 1989. 3.szám [6] Xilinx Foundation Series Quick Start Guide, 1991-1997. Xilinx. Inc.
Downloads
Published
Issue
Section
License
Authors need to sign following agreement with International Journal of MC Square Scientific Research before publishing their articles:
- Authors need to return copyright form to Journal Editor-in-chief to proceed their articles for publication. Meantime, the journal licensed under a Creative Commons Attribution License, which permits other user to distribute the work with an acknowledgement of the authors for International Journal of MC Square Scientific Research.
- Authors are also able to share their separate, additional contractual arrangements for the non-restricted contribution of the journal with an acknowledgement of publication in International Journal of MC Square Scientific Research.
- Authors are allowed and encouraged to share their work during the submission process for increasing citation and exploring to increase the paper availability in worldwide way. The Effect of Open Access.