摘要
为了提高分类精度和边缘辨识性,该文引入图像空间一致性降元(pixels reduction with spatialcoherence property,PRSCP)及线性回归分析,提出了一种基于空间一致性降元的非监督分类。该方法从像元光谱相似性出发,利用像元最小关联窗口合并相邻相似像元为像块完成降元。使用线性关系建模像块内像元的光谱向量,并利用F检验判断像块数据的线性显著性。利用一元线性回归(one dimensional linear regression,ODLR)估计出像块的基准向量,根据基准向量合并相似(同类)像块完成分类。利用AVIRIS数据评估了该方法性能,实验结果表明:与K-MEANS和ISODATA方法相比,该方法精度高、边缘辨识度好及鲁棒性强。
In order to improve classification and edge accuracy, PRSCP and linear regression analysis are introduced; a new algo rithm of unsupervised classification based on PRSCP is proposed. The algorithm procedure starts with the similarity of pixel spectral, and then makes use of minimum related window to combine similar pixels spatially adjacent into a block. Linear cxpres sion is applied to model the spectral vector of pixels in the same block, and significance of the linear expression is verified by F- statistic. The basic vector of block is estimated v/a ODLR, and blocks with similar basic vectors are combined into the same class. AVIRIS data is used to evaluate the performance of the proposed algorithm, which is also compared with K MEANS and ISODATA. Experimental results show that the proposed algorithm outperforms K-MEANS and 1SODATA in terms of classifi- cation accuracy, edge and robustness.
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
2012年第7期1860-1864,共5页
Spectroscopy and Spectral Analysis
基金
国家自然科学基金项目(61071147)资助
关键词
降元
空间一致性
一元线性回归
非监督分类
高光谱图像
Pixels reduction
Spatial coherence property
One dimensional linear regression（ODLR）~ Unsupervised c[assifica-tiom Hyperspectral images
