|
|||||
|
|
| 可变神经网络结构下的遥感影像光谱分解方法 | |
| 引用本文： | 李熙,石长民,李畅,陈锋锐,田礼乔.可变神经网络结构下的遥感影像光谱分解方法[J].计算机工程,2012,38(9):1-3. |
| 作者姓名： | 李熙 石长民 李畅 陈锋锐 田礼乔 |
| 作者单位： | 1. 武汉大学测绘遥感信息工程国家重点实验室,武汉,430079 2. 三亚市国土环境资源信息中心,海南三亚,572000 3. 华中师范大学城市与环境科学学院,武汉,430079 4. 河南大学环境与规划学院,河南开封,475000 |
| 基金项目： | 国家自然科学基金资助项目(41101413);高等学校博士学科点专项科研基金资助项目(20110141120073);中央高校基本科研业务费专项基金资助项目(904275839) |
| 摘 要： | 多层感知神经网络(MLP)是主流的非线性分解方法，但是目前缺乏有效方法处理MLP分解结果中的丰度负值问题。为此，提出一种可变神经网络结构的方法，逐步去除负值丰度对应的端元，并调整相应的网络结构使之针对剩余的端元进行分解。通过武汉地区模拟TM遥感影像实验可以发现，该方法与传统MLP方法以及线性光谱分解方法的平均误差分别为0.077 7、0.081 9、0.094 3，说明该方法的分解精度高于其他2种分解方法，能克服丰度负值问题。 |
| 关 键 词： | 遥感 混合像元 神经网络 多层感知网络 非负约束 非线性光谱分解模型 |
| 收稿时间： | 2011-08-22 |
Spectral Unmixing Method of Remote Sensing Images in Variable Architecture of Neural Network |
|
| LI Xi , SHI Chang-min , LI Chang , CHEN Feng-rui , TIAN Li-qiao.Spectral Unmixing Method of Remote Sensing Images in Variable Architecture of Neural Network[J].Computer Engineering,2012,38(9):1-3. | |
| Authors: | LI Xi SHI Chang-min LI Chang CHEN Feng-rui TIAN Li-qiao |
| Affiliation: | 1(1.State Key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,Wuhan 430079,China;2.Sanya Land Environment and Resources Information Center,Sanya 572000,China;3.College of Urban and Environmental Science,Huazhong Normal University,Wuhan 430079,China;4.College of Environment and Planning,Henan University,Kaifeng 475000,China) |
| Abstract: | Spectral unmixing of remote sensing images is a hotspot in remote sensing field,and Multilayer Perception(MLP) neural network is a common nonlinear spectral unmixing algorithm.However,currently there is no effective way to deal with the negative abundances derived by the network.To solve this problem,a MLP neural network with variable architecture is proposed.By discarding endmembers with negative abundances,the MLP architecture is modified to unmix the rest endmembers,so a remote sensing image is finally unmixed.An experiment using a simulated TM image shows that the average errors of the proposed method,conventional MLP method and linear spectral unmixing model are 0.077 7,0.081 9 and 0.094 3 respectively,thus the proposed method outperforms the other two.Therefore,the proposed method can overcome the negative abundance problem effectively. |
| Keywords: | remote sensing mixed pixel neural network Multilayer Perception(MLP) network nonnegative constraint nonlinear spectral unmixing model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
| 点击此处可从《计算机工程》浏览原始摘要信息 | |
| 点击此处可从《计算机工程》下载全文 | |