基于多重分形参数的高光谱数据特征提取

赵慧洁; 蔡辉; 李娜

基于多重分形参数的高光谱数据特征提取

北京航空航天大学精密光机电一体化技术教育部重点实验室, 北京 100191

基金项目: 国家自然科学基金资助项目(61008047,61177008); 中国地质调查局地质调查资助项目(1212011120227); 长江学者和创新团队发展计划资助项目(IRT0705)

详细信息

中图分类号: TP 775
计量
- 文章访问数: 1860
- HTML全文浏览量: 187
- PDF下载量: 665
- 被引次数: 0
出版历程
- 收稿日期: 2011-12-15
- 网络出版日期: 2012-10-30

Feature extraction method based on multifractal parameters for hyperspectral imagery

Key Laboratory of Precision Opto-Mechatronics Technology, Ministry of Education, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

摘要

摘要: 针对单一分形维数不能表征高光谱数据光谱局部吸收特征的问题,提出了基于光谱概率测度的多重分形参数特征提取方法.基于光谱信息度量进行光谱概率测度的计算,基于配分函数法估计得到尺度函数;通过对尺度函数求导计算出Holder指数,并对尺度函数勒让德Legendre变换计算出多重分形谱;从多重分形谱和Holder指数之间的函数关系提取表征多重分形谱形态的4个多重分形谱参数作为光谱特征参数;并应用于基于最小距离准则的航空推扫式高光谱成像仪(PHI,Prush-broom Hyperspectral Imager)图像监督分类.结果证明:利用基于光谱概率测度的多重分形参数特征提取方法提取的光谱特征参数进行分类得到的总体分类正确率达94.789%,分类精度明显高于利用信息量维数和多重分形谱特征提取方法进行分类的结果,证明了基于光谱概率测度的多重分形参数特征提取方法提取的多重分形参数的有效性和可靠性.
- 高光谱遥感 /
- 特征提取 /
- 多重分形谱参数 /
- 尺度函数 /
- 推扫式高光谱成像仪
Abstract: Multi-fractal parameter extraction method based on spectral probability measurement was proposed to resolve the problem that the local absorption characteristics of hyperspectral data can not be described by the single fractal dimension. The method of spectral information measurement was used to calculate the spectral probability. The scaling function was estimated with the partition function. The differential coefficient of scaling function was calculated to obtain Holder exponent, and the multi-fractal spectrum was computed with Legendre transformation of scaling function. Four multi-fractal parameters can be extracted from multi-fractal spectrum and Holder exponent. The minimum Euclidean distance rule with the characteristic extraction based on multi-fractal parameters was applied to hyperspectral image supervised classification. The hyperspectral image was collected by airborne push-broom hyperspectral imager (PHI). The applied results show that the efficiency and reliability of the proposed method and its classification accuracy are about 94.789%, which is better than the classification accuracy of information fractal dimension and multi-fractal spectrum.
- hyperspectral remote sensing /
- feature extraction /
- multi-fractal parameters /
- scaling function /
- push-broom hyperspectral imager

HTML全文

参考文献(10)

[1]	Zhou Ziyong.Multifractal based hyperion hyperspectral data mining //The Seventh International Conference on Fuzzy System and Knowledge Discovery.USA,NJ:IEEE,2010:2109-2113
[2]	Tzeng Y C,Fan K T,Su Y J,et.al.A parallel differential box counting algorithm applied to hyperspectral image classification //IEEE International Geosciences and Remote Sensing Symposium (IGARSS).USA,NJ:IEEE,2009:V216-V219
[3]	张峰,邹焕新,雷琳.一种基于局部分形维的CFAR检测算法[J].信号处理,2012,28(1):205-211 Zhang Feng,Zou Huanxin,Lei Lin.A CFAR detection algorithm based on local fractal dimension[J].Signal Processing,2012,28(1):205-211(in Chinese)
[4]	Zhang Huaguo,Huang Weigen,Zhou Changbao.Fractal characterization of IKONOS imagery //Multispectral and Hyperspectral Remote Sensing Instruments and Applications.Bellinggham:SPIE,2003,4897:292-301
[5]	周子勇,李朝阳.高光谱遥感数据光谱曲线分形特征研究[J].中北大学学报,2005,26(6):451-454 Zhou Ziyong,Li Chaoyang.Fractal feature of hyperspectral remote sensing spectral curves[J].Journal of North China Institute of Technology,2005,26(6):451-454 (in Chinese)
[6]	Abadi M,Grandchamp E.Legendre spectrum for texture classification //The 8th International Conference on Signal Processing.USA,NJ:IEEE,2006,2:16-20
[7]	Chang Chein I.Spectral information divergence for hyperspectral image analysis //IEEE International Geosciences and Remote Sensing Symposium (IGARSS).USA,NJ:IEEE,1999,1:509-511
[8]	Halsey T C,Jensen M H,Kadanoff L P,et al.Fractal measures and their singularities:the characterization of strange sets [J].Physical Review A,1986,33(2):1141-1151
[9]	周炜星,王延杰,于遵宏.多重分形奇异谱的几何特性Ⅱ.配分函数法[J].华东理工大学学报,2000,26(4):390-395 Zhou Weixing,Wang Yanjie,Yu Zunhong.Geometrical characteristics of singularity spectral of multifractal Ⅱ Partion function definition[J].Journal of East China University of Science and Technology,2000,26(4):390-395(in Chinese)
[10]	李彤,商朋见.多重分形在掌纹识别中的研究[J].物理学报,2007,56(8):4393-4400 Li Tong,Shang Pengjian.A multifractal approach to palmprint recognition[J].Acta Physica Sinica,2007,56(8):4393-4400(in Chinese)