摘要
通过对粗集和神经网络在图像识别中的作用分析,以及对两者结合的可能性研究,将粗集和神经网络进行了有机结合,提出了一个基于粗集和神经网络的图像识别模型。该模型先对原始图像数据进行预处理,然后用粗集进行特征选择,减少了神经网络的输入维数,提高神经网络学习和识别速度,也提高了识别正确率。最后将该模型应用于手写体数字图像识别之中,实验结果表明,该模型是有效的、可行的。
Through analyzing effect of rough set and neural network in image recognition,and researching possibility of combination of both,this paper combines the rough set and neural networks effectively and proposes an image recognition model based on rough set and neural network.The model firstly preprocesses original data of image,then selects features by means of rough set,so it decreases the input dimension of the neural networks and enhances the speed of neural networks studying and distinguishing,and also improves identification rightness rate.Finally,the model is applied to the image of script number classification.The result shows that the method is effective and feasible.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第13期181-184,共4页
Computer Engineering and Applications
基金
四川省科技计划项目(No.2008GZ0003)
四川省科技厅科技攻关项目(No.07GG006-014)
中国科学院人才培养计划项目
关键词
粗集
神经网络
图像识别
rough set
neural network
image recognition
