A Gabor wavelet neural network algorithm was presented based on parallel structure. According to the characteristics of multi- CPU parallel structure and neural network parallel
the algorithm for recognizing image target was designed based on the Gabor wavelet neural network. The algorithm input layer involved operation on scale
translation and frequency modulated parameter of Gabor wavelet
the hide layer realized and optimized the neural network algorithm on parallel CPU and the algorithm output layer gave the the classification result of Gabor wavelet network algorithm. The simulation experiment was processed by four types plane image targets
and the results indicate that the recognition rate is 98%
recognition time is 40 ms.
关键词
Keywords
references
. ZHANG Q,BENVENISTE A. Wavelet networks[J].IEEE Trans. on NN, 1992, 3(11): 889-898.
. PATI Y C,KRISHNAPRASAD P S.Analysis and synthesis of feedforward neural network using discrete affine wavelet[J].IEEE Trans. On NN,1993,4(1):73-75.
. BASKSHI B R,STEPHANOPOULOUS G. Wave-net: a multiresolution, hierarchical neural network with localized learning[J].American Institute Chemical Engineer Journal,1993,39(1):57-81.
. ZHANG J,WALTER G, MIAO Y, et al. Wavelet neural networks for function learning[J]. IEEE Trans.on SP,1995 , 43(6):1485-1497.
. SHI Y, ZHANG X D. Gabor atom networks for signal classification with application in radar target recognition[J]. IEEE Trans. Signal Processing ,2001,49:2994-3004.
. 许廷发. 改进的BP算法在多目标识别中的应用[J]. 光学 精密工程,2003,10(5):,513-515. XU T F. Multil-target recognition with improved BP algorithm[J]. Optics and Precision Engineering,2003(10),513-515.(in Chinese)