根据Ｈｏｐｆｉｅｌｄ神经网络的优化功能设计二元串码滤波器，对于识别同样数目的物体，本方法所需要的滤波器数目较少，且容错能力较强，对适当的编码方法识别准确率可达１００％。
Binary serial-coded-filters (BSCFs) were designed baized on the optimization functions of the Hopfield model. Less filters are heeded to carry out the same recognition task, compared with other methods. The error tolerant capacity is also very strong. All target objects can be correctly recognized when the characteristic codes are properly chosed.
刘颖 路明哲.用Hopfield网络实现二元串码滤波器[J].红外与毫米波学报,1994,13(3):221~226]. Liu Ying, Lu Mingzhe, Zhang Jianming Fang Zhiliang, Liu Fulai, Mu Guoguang. DESIGN OF BINARY SERIAL-CODED-FILTERS BY USING HOPFIELD MODEL[J]. J. Infrared Millim. Waves,1994,13(3):221~226.]
复制
