当前位置: 首页 > 文章 > 人工神经网络在农村土地利用分类中的应用 农机化研究 2011,33 (1) 190-194
Position: Home > Articles > Application of Artificial Neural Networks in Village Land Use Classification Journal of Agricultural Mechanization Research 2011,33 (1) 190-194

人工神经网络在农村土地利用分类中的应用

作  者:
郭小英;何东健
单  位:
西北农林科技大学机械与电子工程学院
关键词:
土地利用;遥感图像分类;人工神经网络
摘  要:
为克服传统方法在土地利用分类中的不足,提出了以Google Earth公开遥感图像为样本,在采用灰度共生矩阵方法提取图像纹理特征和利用主成分分析法进行特征优选的基础上,建立BP神经网络图像分类的遥感图像土地利用分类模型。以Matlab工具为平台对实验图像进行验证。结果表明:该分类模型分类总体精度达到88.00%,Kappa系数达到0.814 5,优于传统的最大似然分类方法,对农村资源规划与环境调查有较大帮助。
译  名:
Application of Artificial Neural Networks in Village Land Use Classification
作  者:
Guo Xiaoying,He Dongjian(College of Mechanical and Electronic Engineering,Northwest Agriculture & Forest University,Yangling 712100,China)
关键词:
land use classification;remote sensing image classification;artificial neural networks
摘  要:
To overcome the traditional method's drawbacks in land use classification google earth images were used as sample images to make the experiments.Firstly,the different features were extracted in the images by gray-level co-occurrence Matrix method and the useful features were selected by principal component analysis method,then,the regions of interest in images were drawn for classifying the sample features,finally,the BP network model was founded for images classification.The experiments results of images based on Matlab platform show that the accuracy of BP network method is up to 88.00% and Kappa is 0.8145,which is superior to traditional maximum likelihood method and is helpful for land resources and environment survey.

相似文章

计量
文章访问数: 8
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊