基于稀疏表示的多光谱与全色遥感图像融合新方法
|
摘要
针对多光谱图像和全色图像融合问题,提出了一种基于稀疏表示的多光谱与全色遥感图像融合新方法。首先对多光谱图像进行IHS变换,得到I亮度分量,然后采用滑动窗口技术对I亮度分量和全色图像分块,利用过完备字典和正交匹配算法(OMP)得到I亮度分量和全色图像的稀疏系数,采用Sigmoid函数对稀疏系数进行融合。文中使用Quick Bird卫星数据验证本文提出方法的有效性。实验结果表明,相比于传统方法,本文方法得到的融合图像具有较好的光谱特性,保留了更多的空间细节信息。
A new fusion method of multi-spectral image and panchromatic remote sensing image base on sparse representation is proposed as follows. Firstly, the multi-spectral image is converted to IHS, which produces an array of luma samples, and then the sliding window technique is used to divide the panchromatic image and one array of luma samples into pieces. The complete dictionary is combined with OMP algorithm to extract sparse coefficients for the panchromatic image and one array of luma samples. Sigmoid function is used to fuse the sparse coefficients. QuickBird satellite data verify the effectiveness of the above method. The proposed method can keep better spectral properties and more spatial information of the fused image in comparison with traditional method.
A new fusion method of multi-spectral image and panchromatic remote sensing image base on sparse representation is proposed as follows. Firstly, the multi-spectral image is converted to IHS, which produces an array of luma samples, and then the sliding window technique is used to divide the panchromatic image and one array of luma samples into pieces. The complete dictionary is combined with OMP algorithm to extract sparse coefficients for the panchromatic image and one array of luma samples. Sigmoid function is used to fuse the sparse coefficients. QuickBird satellite data verify the effectiveness of the above method. The proposed method can keep better spectral properties and more spatial information of the fused image in comparison with traditional method.
引文
[1]Choi M.A new intensity-hue-saturation fusion approach to image fusion with a tradeoff parameter[J].IEEETransactions on Geoscience and Remote sensing,2006,44(6):1672-1682.
[2]李振华,敬忠良,孙韶媛,等.基于方向金字塔框架变换的遥感图像融合算法[J].光学学报,2005,25(5):598-602.
[3]余连生,文贡坚,李智勇.基于SIDWT的遥感图像融合算法[J].计算机工程,2011,37(17):197-199.
[4]Hu J,Li S,Yang B.Remote sensing image fusion based on IHS transform and sparse representation[C]//Proceedings of Chinese Conference on Pattern Recognition.Chongqing,China:IEEE,2010:1-4.
[5]张晓,薛月菊,涂淑琴,等.基于结构组稀疏表示的遥感图像融合[J].中国图象图形学报,2016,21(8):1106-1118.
[6]朱超.多光谱与全色遥感图像融合算法的研究[D].合肥:安徽大学,2013.
[7]张健.基于稀疏表示模型的图像复原技术研究[D].哈尔滨:哈尔滨工业大学,2014.
[2]李振华,敬忠良,孙韶媛,等.基于方向金字塔框架变换的遥感图像融合算法[J].光学学报,2005,25(5):598-602.
[3]余连生,文贡坚,李智勇.基于SIDWT的遥感图像融合算法[J].计算机工程,2011,37(17):197-199.
[4]Hu J,Li S,Yang B.Remote sensing image fusion based on IHS transform and sparse representation[C]//Proceedings of Chinese Conference on Pattern Recognition.Chongqing,China:IEEE,2010:1-4.
[5]张晓,薛月菊,涂淑琴,等.基于结构组稀疏表示的遥感图像融合[J].中国图象图形学报,2016,21(8):1106-1118.
[6]朱超.多光谱与全色遥感图像融合算法的研究[D].合肥:安徽大学,2013.
[7]张健.基于稀疏表示模型的图像复原技术研究[D].哈尔滨:哈尔滨工业大学,2014.