基于多元遥感影像分割和区域特征相似度的微波土壤水分反演靶区选择方法
|
摘要
为解决微波遥感土壤水分反演经验方程(关系)的适用性问题,该文提出一个新的概念——反演靶区,并提出了一种基于多元遥感影像分割和区域特征相似度的微波土壤水分反演靶区选择方法。首先采用主成分分析法(PCA),从影响土壤含水量的12个因子(土壤湿度分布、地表温度、NDVI、土壤质地指数、地形粗糙度、雷达入射角以及Landsat TM除热红外波段的6个波段)中筛选提取第一主成分;再使用Mean Shift方法分割包含第一主成分的单色影像,得到一幅过分割区域图;计算各分割区域与各样方区域的特征向量间的加权欧氏距离,得到区域特征相似度数据集;最后,基于各样方得到的反演经验方程(关系),根据区域相似度选择各自反演靶区进行土壤水分反演。实验证明,与只使用同一反演经验方程进行反演相比,基于反演靶区思想的土壤水分反演精度有较大提高。
Traditionally,a retrieving empirical equation or relationship is developed based on the fitting or training between measured data of sampling points and SAR data at the same time.Since the range and number of sampling quadrats are limited,there should be a difference between sampling quadrats and other areas.Therefore,for one retrieving empirical equation or relationship,there is only a suitable area,and the area is named retrieving target areas in this paper.In order to solve this problem,a new method for choosing retrieving target areas during soil moisture inversion using microwave based on multivariate remote sensing image segmentation and region feature similarity is proposed.First,the Principal Component Analysis(PCA)was used to extract the first principal component from 12 factors affecting soil moisture.Then,the monochrome image containing the first principal component was segmented based on the Mean Shift method.A graph of a split target area could be got.A data set about regional feature similarity was obtained by calculating the weighted Euclidean distance between 6 dimensional feature vectors of each segmentation region and each quadrat area.These components of the region feature vector included soil moisture,land surface temperature,NDVI,soil texture index,surface roughness and radar incidence angle.At last,each retrieving empirical equation or relationship was used for soil moisture inversion by choosing their retrieving target areas based on regional feature similarity.To prove the validity of this method,the results of this method were compared with the results of soil moisture inversion by using only one empirical equation.Results indicated that the accuracy of soil moisture inversion based on retrieving target areas was greatly improved.
Traditionally,a retrieving empirical equation or relationship is developed based on the fitting or training between measured data of sampling points and SAR data at the same time.Since the range and number of sampling quadrats are limited,there should be a difference between sampling quadrats and other areas.Therefore,for one retrieving empirical equation or relationship,there is only a suitable area,and the area is named retrieving target areas in this paper.In order to solve this problem,a new method for choosing retrieving target areas during soil moisture inversion using microwave based on multivariate remote sensing image segmentation and region feature similarity is proposed.First,the Principal Component Analysis(PCA)was used to extract the first principal component from 12 factors affecting soil moisture.Then,the monochrome image containing the first principal component was segmented based on the Mean Shift method.A graph of a split target area could be got.A data set about regional feature similarity was obtained by calculating the weighted Euclidean distance between 6 dimensional feature vectors of each segmentation region and each quadrat area.These components of the region feature vector included soil moisture,land surface temperature,NDVI,soil texture index,surface roughness and radar incidence angle.At last,each retrieving empirical equation or relationship was used for soil moisture inversion by choosing their retrieving target areas based on regional feature similarity.To prove the validity of this method,the results of this method were compared with the results of soil moisture inversion by using only one empirical equation.Results indicated that the accuracy of soil moisture inversion based on retrieving target areas was greatly improved.
引文
[1]赵英时.遥感应用分析原理与方法[M].北京:科学出版社,2014.
[2]王东.基于SAR图像的植被覆盖下土壤含水量反演方法研究[D].成都:电子科技大学,2014.
[3]邓书斌.ENVI遥感图像处理方法[M].北京:科学出版社,2014.
[4]周家香.Mean Shift遥感图像分割方法与应用研究[D].长沙:中南大学,2012.
[5]刘英,吴立新,马保东.基于TM/ETM+光谱特征空间的土壤湿度遥感监测[J].中国矿业大学学报,2013,42(2):296-301.
[6]覃志豪,ZHANG M H,KARNIELI A,等.用陆地卫星TM6数据演算地表温度的单窗算法[J].地理学报,2001,56(4):456-466.
[7]乔平林,张继贤,燕琴,等.利用TM6进行土壤水分的监测研究[J].测绘通报,2003(7):494-911.
[8]洪奕丰,严恩萍,林辉,等.浙东地貌形态与土地覆被格局关系的研究[J].中南林业科技大学学报,2012,32(3):63-69.
[9]王建社.SAR图像入射角效应校正与分割方法研究[D].合肥:合肥工业大学,2011.
[10]FUKUNAGA K,HOSTETLER L D.The estimation of the gradient of a density function,with applications in pattern recognition[J].IEEE Transactions on Information Theory,1975,21(1):32-40.
[11]HUANG J,KUMAR S R,MITRA M,et al.Spatial color indexing and applications[J].International Journal of Computer Vision,1999,35(3):245-268.
[12]梁慧琳.基于颜色纹理特征的均值漂移图像分割改进算法研究[D].银川:宁夏大学,2013.
[13]庄越挺,潘云鹤,吴飞.网上多媒体信息分析与检索[M].北京:清华大学出版社,2002.
[14]XIANG S,NIE F,ZHANG C.Learning a Mahalanobis distance metric for data clustering and classification[J].Pattern Recognition,2008,41(12):3600-3612.
[15]BUADES A,COLL B,MOREL J M.A non-local algorithm for image denoising[A].IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR′05)[C].2005.
[16]DOBSON M C,ULABY F T,HALLIKAINEN M T,et al.Microwave dielectric behavior of wet soil[J].IEEE Transactions on Geoscience and Remote Sensing,1985,23(1):35-46.
[17]宋宇晨,张玉英,孟海东.一种基于加权欧式距离聚类方法的研究[J].计算机工程与应用,2007,43(4):179-180.
[18]陈鲁皖.基于GIS的活断层灾害危险性评价——以海原活动断裂带为例[D].西安:长安大学,2007.
[19]SAATY T.The analytic hierarchy process[M].New York:McGraw-Hill Inc,1980.
[20]SHI J C,WANG J,NEILL P,et al.Estimation of bare surface soil moisture and surface roughness parameter using L-band SAR image data[J].IEEE Transactions on Geoscience and Remote Sensing,1997,35(5):1254-1266.
[21]余凡,赵英时.合成孔径雷达反演裸露地表土壤水分的新方法[J].武汉大学学报(信息科学版),2010,35(3):317-321.
[22]陈晶,贾毅,余凡.双极化雷达反演裸露地表土壤水分[J].农业工程学报,2013,29(10):109-116.
[23]孔金玲,甄珮珮,李菁菁,等.基于新的组合粗糙度参数的土壤水分微波遥感反演[J].地理与地理信息科学,2016,32(3):34-38.
[24]王军战,张友静,鲍艳松,等.基于ASAR双极化雷达数据的半经验模型反演土壤湿度[J].地理与地理信息科学,2009,25(2):5-9.
[25]LIEVENS H,VERHOEST N E C,KEYSER E D,et al.Effective roughness modelling as a tool for soil moisture retrieval from C-and L-band SAR[J].Hydrology and Earth System Sciences,2011,15(1):151-162.
[2]王东.基于SAR图像的植被覆盖下土壤含水量反演方法研究[D].成都:电子科技大学,2014.
[3]邓书斌.ENVI遥感图像处理方法[M].北京:科学出版社,2014.
[4]周家香.Mean Shift遥感图像分割方法与应用研究[D].长沙:中南大学,2012.
[5]刘英,吴立新,马保东.基于TM/ETM+光谱特征空间的土壤湿度遥感监测[J].中国矿业大学学报,2013,42(2):296-301.
[6]覃志豪,ZHANG M H,KARNIELI A,等.用陆地卫星TM6数据演算地表温度的单窗算法[J].地理学报,2001,56(4):456-466.
[7]乔平林,张继贤,燕琴,等.利用TM6进行土壤水分的监测研究[J].测绘通报,2003(7):494-911.
[8]洪奕丰,严恩萍,林辉,等.浙东地貌形态与土地覆被格局关系的研究[J].中南林业科技大学学报,2012,32(3):63-69.
[9]王建社.SAR图像入射角效应校正与分割方法研究[D].合肥:合肥工业大学,2011.
[10]FUKUNAGA K,HOSTETLER L D.The estimation of the gradient of a density function,with applications in pattern recognition[J].IEEE Transactions on Information Theory,1975,21(1):32-40.
[11]HUANG J,KUMAR S R,MITRA M,et al.Spatial color indexing and applications[J].International Journal of Computer Vision,1999,35(3):245-268.
[12]梁慧琳.基于颜色纹理特征的均值漂移图像分割改进算法研究[D].银川:宁夏大学,2013.
[13]庄越挺,潘云鹤,吴飞.网上多媒体信息分析与检索[M].北京:清华大学出版社,2002.
[14]XIANG S,NIE F,ZHANG C.Learning a Mahalanobis distance metric for data clustering and classification[J].Pattern Recognition,2008,41(12):3600-3612.
[15]BUADES A,COLL B,MOREL J M.A non-local algorithm for image denoising[A].IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR′05)[C].2005.
[16]DOBSON M C,ULABY F T,HALLIKAINEN M T,et al.Microwave dielectric behavior of wet soil[J].IEEE Transactions on Geoscience and Remote Sensing,1985,23(1):35-46.
[17]宋宇晨,张玉英,孟海东.一种基于加权欧式距离聚类方法的研究[J].计算机工程与应用,2007,43(4):179-180.
[18]陈鲁皖.基于GIS的活断层灾害危险性评价——以海原活动断裂带为例[D].西安:长安大学,2007.
[19]SAATY T.The analytic hierarchy process[M].New York:McGraw-Hill Inc,1980.
[20]SHI J C,WANG J,NEILL P,et al.Estimation of bare surface soil moisture and surface roughness parameter using L-band SAR image data[J].IEEE Transactions on Geoscience and Remote Sensing,1997,35(5):1254-1266.
[21]余凡,赵英时.合成孔径雷达反演裸露地表土壤水分的新方法[J].武汉大学学报(信息科学版),2010,35(3):317-321.
[22]陈晶,贾毅,余凡.双极化雷达反演裸露地表土壤水分[J].农业工程学报,2013,29(10):109-116.
[23]孔金玲,甄珮珮,李菁菁,等.基于新的组合粗糙度参数的土壤水分微波遥感反演[J].地理与地理信息科学,2016,32(3):34-38.
[24]王军战,张友静,鲍艳松,等.基于ASAR双极化雷达数据的半经验模型反演土壤湿度[J].地理与地理信息科学,2009,25(2):5-9.
[25]LIEVENS H,VERHOEST N E C,KEYSER E D,et al.Effective roughness modelling as a tool for soil moisture retrieval from C-and L-band SAR[J].Hydrology and Earth System Sciences,2011,15(1):151-162.