摘要
焙面包切片区域图像的灰值游程矩阵提取数学统计量参数作为纹理特征,利用神经网络实现对面包品质的分类。由于神经网络是一种黑箱操作,难以对分布在其中的知识进行解释。采用基于对隐层神经元输出值聚类的遗传算法实现了对面包品质分类的规则抽取,实验结果表明该方法具有优良的识别效果。
The method of the baking quality-test of bread classification based on the text features extracted according to the gray run matrix on FOV （field of view） of baked bread slices is described. Because neural network is a black box model, it is difficult to explain the knowledge hidden in it. Clustering genetic algorithm based on the clustering of the hidden units activation values is used to extract the rules. The experiment results indicate the good performance.
出处
《计算机工程与设计》
CSCD
北大核心
2007年第23期5767-5769,共3页
Computer Engineering and Design
基金
国家"粮食丰产科技工程"基金项目(2004BA520A)
