节点文献
神经网络模式识别系统互连权重二值化研究
Study on Binary Interconnection Weight ofa Cascaded Neural Network for Pattern Recognition System
【摘要】 在增量算法的基础上,利用截断(Clipping)方法和蒙塔卡罗(MonteCarlo)算法,对以四类飞行目标平面旋转投影作为学习样本的级联神经网络互连权重进行了二值优化处理,并用非学习样本进行了容错性检验,计算机模拟得到了满意的结果
【Abstract】 In this paper, based on the increment algorithm, the clipping learning method and Monto Carlo algorithm were used in optimization of a cascaded neural network. As a result, binary interconnection weights were obtained. The error tolerance of the neural network was tested by non learning sets. Computer simulation indicated that the results were satisfactory.
【关键词】 模式识别系统;
神经网络互连权重;
灰度阶;
二值化;
【Key words】 pattern recognition system; interconnection weight; gray levels; binary.;
【Key words】 pattern recognition system; interconnection weight; gray levels; binary.;
【基金】 国家自然科学基金和攀登计划所资助
- 【文献出处】 光学学报 ,Acta Optica Sinica , 编辑部邮箱 ,1996年10期
- 【分类号】TP183
- 【被引频次】4
- 【下载频次】90