一种基于最大距离的纯像元指数端元提取算法
作者简介:徐君(1979-),男,汉族,山东济南人,博士,讲师,研究方向为光谱成像、高光谱遥感。E-mail:xjsdcq@163.com
收稿日期: 2013-12-09
要求修回日期: 2014-03-05
网络出版日期: 2015-01-05
基金资助
国家自然科学青年基金项目(41301382)
教育部人文社会科学研究青年基金项目(14YJCZH172、13YJCZH089)
江西省自然科学基金项目(20122BAB211040、20142BAB207007)
江西省教育厅科学技术研究项目(GJJ14368)
A Novel Pure Pixel Index Endmember Extraction Algorithm Based on the Maximum Distance
Received date: 2013-12-09
Request revised date: 2014-03-05
Online published: 2015-01-05
Copyright
在高光谱混合像元分解中,PPI算法是一种比较成熟的算法,但PPI算法中每次投影向量的生成都是随机的,多次执行PPI算法后端元提取的结果并不稳定。本文以线性光谱混合模型的凸面几何学描述为基础,利用端元在高光谱图像特征空间中所形成的凸面单形体端点的特点,提出了一种区别于PPI算法的最大距离纯像元指数方法。选取特征空间中所有样本点的光谱均值作为超球的球心,计算所有样本点到球心的欧氏距离,以等于或大于这个最大距离的长度作为半径,在特征空间中设计一个包围所有样本点的超球面,并在超球面上均匀地选取参考点,针对每一个参考点,在样本点中找出与它距离最远的一个,记录每个样本点成为距离最大点的次数,将其作为评价该像元是否为端元的纯像元指数,从而使得每次端元提取的精度得到保证。最后,利用美国内华达州Cuprite获取的AVIRIS数据对算法进行了验证。实验结果表明,采用本文算法提取的端元精度优于N-FINDR算法和VCA算法,而且鲁棒性较好,克服了PPI算法由于随机生成投影向量所带来的端元提取不稳定性。
徐君 , 徐富红 , 蔡体健 , 王彩玲 , 黄德昌 , 李伟平 . 一种基于最大距离的纯像元指数端元提取算法[J]. 地球信息科学学报, 2015 , 17(1) : 86 -90 . DOI: 10.3724/SP.J.1047.2015.00086
In hyperspectral unmixing, PPI algorithm is a relatively mature algorithm, but each projection vector in PPI algorithm is generated randomly, and the endmembers extracted by PPI algorithm are not stable. That is, different endmembers can be obtained from the same image by repeatedly running PPI algorithm. This paper, based on the convex geometry description of linear spectral mixing model, utilized the feature that the endmembers are the endpoints of the single convex body enclosed in the hyperspectral image feature space, and proposed a novel pure pixel index algorithm for endmember extraction based on the maximum distance. The average of the spectral vectors of all the sample points is calculated and used as the center of a hypersphere. Next, we calculate the Euclidean distances of all the sample points to the center of the hypersphere, and design a radius of equal to or greater than the maximum distance for the hypersphere in the feature space to include all of the sample points. We evenly select the reference points on the surface of the hypersphere. The farthest sample point with respect to each reference point can be found by calculating the Euclidean distance. Subsequently, every sample point’s frequency of being the most distant to the reference points is recorded as an index to evaluate whether the sample point is an endmember or not. Finally, we use the AVIRIS data of Nevada Cuprite to testify this algorithm. The experimental results illustrate that the precision of the endmember extraction using the algorithm proposed in this paper is better than N-FINDR algorithm and VCA algorithm in general. Moreover, it has a good robustness and could overcome the instability of PPI algorithm caused by random projection.
Key words: hyperspectral; mixed pixel; endmember extraction; maximum distance; pure pixel index
Fig. 1 The schematic diagram of endmember extraction algorithm based on maximum distance pure pixel index图1 最大距离纯像元指数提取端元的原理示意图 |
Fig. 2 The schematic diagram of reference points selection based on hypersphere图2 超球面上选择参考点的原理示意图 |
Fig. 3 The false color composite image of the AVIRIS data of Cuprite area图3 Cuprite区域的AVIRIS数据假彩色合成图 |
Fig. 4 The pure pixel images obtained by the algorithm proposed in this paper and by the PPI algorithm图4 本文算法及PPI算法生成的像元纯净影像 |
Fig. 5 The histograms obtained from two pure pixel images图5 像元纯净影像的灰度直方图 |
Tab. 1 The contrast of the algorithm proposed in this paper with N-FINDR and VCA表1 本文算法与N-FINDR和VCA的对比 |
矿物 | 本文算法(50次均值) | PPI(50次均值) | VCA | N-FINDR |
---|---|---|---|---|
皂石(Nortronite) | 0.0568 | 0.0836 | 0.0713 | 0.0576 |
明矾石(Alunite) | 0.0622 | 0.0829 | 0.0862 | 0.0635 |
蒙脱石(Montmorillonite) | 0.0626 | 0.0743 | 0.0642 | 0.0961 |
玉髓(Chalcedony) | 0.1048 | 0.1512 | 0.1243 | 0.1219 |
沙漠地表(Desert Varnish) | 0.1022 | 0.1141 | 0.1081 | 0.1131 |
铵长石(Buddingtonite) | 0.0939 | 0.1254 | 0.1056 | 0.0916 |
白云母(Muscovite) | 0.0706 | 0.0982 | 0.0751 | 0.0808 |
高岭石(Kaolinite) | 0.1295 | 0.1442 | 0.1348 | 0.1303 |
榍石(Sphene) | 0.0612 | 0.0813 | 0.0627 | 0.0726 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
/
〈 | 〉 |