遥感图像识别中粗糙集理论与神经网络的结合

Combining Rough Set Theory with Neural Network for Remote Sensing Image

由于传统神经网络与遥感图像信息量不相匹配 ,为此 ,提出将粗糙集理论集成至遥感图像神经网络识别中。首先分析了神经网络与粗糙集理论结合的可能性以及优势 ,在此基础上提出了基于粗糙集的遥感图像神经网络识别模型 ,并就其中的粗糙集方法处理样本特征集模块和遥感图像识别神经网络模块展开详细的分析。通过对比实验数据说明集成粗糙集理论的遥感图像神经网络识别能够有效提高遥感图像的识别效率 。

There is a gap between traditional neural network's information process ability and amount information of remote sensing image recognition. Focusing on this problem, this paper proposes a method to combine rough set theory with neural network theory and uses it in remote sensing image recognition. First, this paper analyzes the feasibility and advantages of combination of neural network with rough set theory. Based on this analysis, a rough set theory based remote sensing image recognition model is presented. Furthermore, analysis on rough set module and remote sensing image recognition module are given in details. Finally, contrastive experiment data are given to prove that combining rough set theory with neural network theory for remote sensing image recognition has a high converging rate, shorter training time and more accuracy.The potential of this method,is also shown.