利用离散Hopfield神经网络对手写数字识别进行了研究。将受到噪声污染的手写数字储存为二值图像,然后调制成二进制信号通过神经网络进行传输,通过给定权矩阵的Hopfield神经网络进行按址存储,将网络输出的内容再映射为数字图像。实验结果表明:数字图像识别的误码率与调制信号的幅值、码间时间间隔和网络神经元耦合个数成负相关关系,而且随着噪声强度的增加,误码率出现非周期随机共振现象,在一非零最优噪声强度值达到最小,此时数字图像也恢复得更加清晰。这些结果为进一步研究最小误码率优化目标下的Hopfield神经网络自适应权重矩阵提供了实验依据,而且对于神经网络联想记忆中随机因素的作用研究具有重要意义。
This paper studies the handwritten digital recognition by the discrete Hopfield neural networks. In the experiment, the noisy handwritten digital image is transferred into the binary signal by the serial-scan mode. The binary modulated signal is transmitted through the neural network with the designed weight matrix and the output storage mode of the network is mapped into digital image. The error rate of the digital image is negatively correlated with the amplitude of the modulated signal, the time interval and the number of coupled neurons in the network. However, as the noise intensity increases, the error rate manifests the aperiodic stochastic resonance effect, and achieves the minimum at the non-zero optimal noise intensity. Under this circumstance, the recovered digital image appears more clearly. These results provide a theoretical basis for further research on the adaptive weight matrix of Hopfield neural network for obtaining the minimum error rate, and also are of significance for the positive role of the randomness in the associative memory of the neural networks.
[1]Collins J J, Chow C C, Capela A C, Aperiodic stochastic resonance[J]. Physical Review E, 1996, 54(5): 5575.
[2]Warwick K. Neural networks: an introduction[M]. IET Digital Library, 1992.
[3]王强,张小溪,韩一红. 基于神经网络的图像识别[J]. 电子设计工程, 2012, 20(9):187189.
Wang Qiang, Zhang Xiaoxi, Han Yihong, et al. Image recognition based on artificial neural network[J]. Electronic Design Engineering, 2012, 20(9): 187189.
[4]Crisanti A, Falcioni M, Paladin G, et al. Stochastic resonance in deterministic chaotic systems [J]. Journal of Physics A, 1994, 27(17) : L597L603.
[5]朱献文. 基于遗传算法和Hopfield神经网络的字符识别方法[J]. 电子设计工程, 2011, 19(18): 5759.
Zhu Xianwen. Character recognition method based on genetic algorithm and Hopfield Neural Network [J]. Electronic Design Engineering, 2011, 19(18): 5759.
[6]魏武,黄心汉,张起森,等. 基于模板匹配和神经网络的车牌字符识别方法[J]. 模式识别与人工智能,2001, 14(1): 123127.
Wei Wu, Huang Xinhan, Zhang Qisen, et al. A Method of recognizing characters in vehicle license plates using pattern match and neural networks[J]. Pattern Recognition and Artificial Intelligence, 2001, 14(1): 123127.
[7]丘敏,麦汉荣,廖惜春. 神经网络在汽车牌照字符识别中的应用[J]. 计算机工程与设计,2008, 29(8): 20412043.
Qiu Min, Mai Hanrong, Liao Xichun. Application of neural network in vehicle license character recognition[J]. Computer Engineering and Design, 2008, 29(8): 20412043.
[8]Tatem A J, Lewis H G, Atkinson P.M, et al. Super-resolution target identification from remotely sensed images using a Hopfield neural network[J]. IEEE Transactions on Geoscience & Remote Sensing, 2001, 39(4): 781796.
[9]Katada N, Nishimura H. Stochastic resonance in recurrent neural network with Hopfield-Type memory[J]. Neural Processing Letters, 2009, 30(2): 145154.
[10] Pinamonti G, Marro J, Torres J J. Stochastic resonance crossovers in complex networks.[J]. PloS ONE, 2012, 7(12): e51170.
[11] Nishimura H, Katada N, Aihara K. Deterministic SR phenomena in autoassociative chaotic neural networks[C]. International Conference on Neural Information Processing, IEEE, Singapore, 2002, 2: 585589.
[12] Hopfield J J. Neural networks and physical systems with emergent collective computational abilities[J]. PNAS, 1982, 79(8): 25542558.
[13] Haykin S. Neural Networks and Leaning Machines [M]. Beijing: China Machine Press, 2017.
[14] Gabbiani F, Cox J S. Mathematics for Neuroscientists[M]. Beijing:Academic Press,2012.
[15] Aiyer S V B,Niranjan M,Fallside F. A theoretical investigation into the performance of the Hopfield model[M]. New York: IEEE Press, 1990.
[16] Gonzalez R C, Woods R E, Eddins S L, et al. Digital Image Processing Using MATLAB [M]. Beijing: Publishing House of Electronics Industry, 2013.
[17] McDonnell M D, Abbott D. What is stochastic resonance? Definitions, misconceptions, debates, and its relevance to biology [J]. PLoS Computational Biology, 2009, 5(5): e1000348.
[18] Duan F, Abbott D. Binary modulated signal detection in a bistable receiver with stochastic resonance[J]. Physica A: Statistical Mechanics & Its Applications, 2007, 376(1): 173190.
[19] Riani M,Simonotto E. Stochastic resonance in the perceptual interpretation of ambiguous figures: a neural network model[J]. Physical Review Letters, 1994, 72(19): 3120.
[20] Wiesenfeld K, Moss F. Stochastic resonance and the benefits of noise: from ice ages to crayfish and Squids [J]. Nature, 1995, 373(6509): 3336.
