|
|||||
|
|
| 稀疏约束下的MVC-NMF算法研究 | |
| 引用本文： | 李二森,邹瑜,苏斌. 稀疏约束下的MVC-NMF算法研究[J]. 海洋测绘, 2011, 31(1): 43-46 |
| 作者姓名： | 李二森 邹瑜 苏斌 |
| 作者单位： | 1. 解放军信息工程大学测绘学院,河南,郑州,450052;武汉大学测绘遥感信息工程国家重点实验室,湖北,武汉,430079 2. 解放军信息工程大学测绘学院,河南,郑州,450052 3. 75719部队,湖北,武汉,430047 |
| 摘 要： | 最小体积约束的非负矩阵分解算法(MVC-NMF)不需要假定纯像元的存在,并且在自动提取端元的同时,能够获取每种端元所对应的丰度图,然而该方法并没有考虑丰度矩阵的稀疏特性。提出了将稀疏约束引入MVC-NMF算法中,提高了算法的精度。实验结果表明:该方法在相同迭代次数条件下比MVC-NMF算法解混的精度更高。 |
| 关 键 词： | 混合像元 端元 丰度 稀疏约束 非负矩阵分解 |
Research of the Sparseness Constraint MVC-NMF Algorithm |
|
| LI Er-sen,ZOU Yu,SU Bin. Research of the Sparseness Constraint MVC-NMF Algorithm[J]. Hydrographic Surveying and Charting, 2011, 31(1): 43-46 | |
| Authors: | LI Er-sen ZOU Yu SU Bin |
| Affiliation: | LI Er-sen1,2,ZOU Yu1,SU Bin3(1.Institute of Surveying and Mapping,Information Engineering University,Zhengzhou,Henan,450052,2.State key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,Wuhan,Hubei,430079,3.75719 Troops,430079) |
| Abstract: | The mixels in the hypersepectral images directly influence the accuracy of target recognition,spectral unmixing can effectively solve this problem.The minimum volume constraint nonnegative matrix factorization(MVC-NMF) algorithm is not based on the hypothesis of the pure pixels' existence,and can get the abundance maps as soon as the endmembers are all extracted,but it doesn't take account the sparseness of the abundance matrix,and affects the unmixing precision.We propose to use the sparse constraint into ... |
| Keywords: | mixel endmember abundance sparse constraint nonnegative matrix factorization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
| 点击此处可从《海洋测绘》浏览原始摘要信息 | |
| 点击此处可从《海洋测绘》下载全文 | |