DocumentCode :
2341065
Title :
A multiplier-less GA optimized pulsed neural network for satellite image analysis using a FPGA
Author :
Hualiang Zhuang ; Low, Kay-Soon ; Yau, Wei-Yun
Author_Institution :
Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore
fYear :
2008
fDate :
3-5 June 2008
Firstpage :
302
Lastpage :
307
Abstract :
This paper presents a digital hardware oriented system that uses a genetic algorithm (GA) for optimizing a pattern classifier based on the pulsed neural network (PNN). The scheme avoids the usage of multipiers and dividers, which are the bottlenecks for digital hardware implementation of parallel computations like GA and neural networks. Utilizing the nature of RBF being inherent in the pulsed neural network, the scheme yields very compact computational circuits for implementation on a FPGA chip with massive parallelism that guarantees the speed of the neural and evolutionary computations. The on-chip GA-PNN system is developed for terrain classification of a multi-spectral satellite image. Experimental results show that the performance of the proposed system is comparable to a back propagation (BP) neural network while its training speed exceeds the BP network overwhelmingly.
Keywords :
artificial satellites; field programmable gate arrays; genetic algorithms; geophysics computing; image classification; parallel processing; radial basis function networks; terrain mapping; FPGA; compact computational circuits; digital hardware oriented system; evolutionary computations; multiplier-less genetic algorithm; parallel computations; pattern classifier optimisation; pulsed neural network; radial basis function; satellite image analysis; terrain classification; Computer networks; Concurrent computing; Field programmable gate arrays; Genetic algorithms; Image analysis; Neural network hardware; Neural networks; Parallel processing; Pulse circuits; Satellites;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Industrial Electronics and Applications, 2008. ICIEA 2008. 3rd IEEE Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-1717-9
Electronic_ISBN :
978-1-4244-1718-6
Type :
conf
DOI :
10.1109/ICIEA.2008.4582529
Filename :
4582529
Link To Document :
بازگشت