基于改进的SLIC和OTSU的遥感影像水体提取
|
摘要
水体信息提取是水资源管理的重要组成部分,近二十年来关于从遥感图像中提取水体信息的研究有许多。基于卫星遥感的陆地水体提取方法多种多样,利用水体指数的OTSU算法就是一种较为常见的水体提取方法,但是其存在阈值选取困难和提取精度不足的问题。为了进一步解决该问题,将改进的SLIC和OTSU算法引入到水体信息的提取中。考虑到水体指数对于水体信息提取的重要性,首先利用超像素生成算法SLIC结合归一化水体指数(NDWI)生成一个个超像素,以超像素内水体指数值的均值代表这个超像素的水体指数。然后以超像素为基础,使用OTSU算法对超像素水体指数做阈值分割。实验结果表明,提出的算法不仅有很高的提取精度,提取的速度也很快,能够有效地提取出水体信息。
Water information extraction is an important part of water resources management. There have been many researches on water information extraction from remote sensing images in the past two decades. There are a variety of land-based water extraction methods based on satellite remote sensing. OTSU algorithm using water index is a relatively common one,but it is difficult to select the threshold and the extraction accuracy is insufficient. For this,we introduce the improved SLIC and OTSU algorithm into water information extraction.Considering the importance of water index for water information extraction,we combine SLIC,a super-pixel generation algorithm,and NDWI,the normalized water index,to generate each super-pixel,and the mean value of water index within the super-pixel represents the water index of this super-pixel. Then based on super-pixel,OTSU algorithm is used to conduct the threshold segmentation of the water index in the super-pixel. Experiment shows that the proposed algorithm not only has a high extraction accuracy,but also a fast extraction speed,which can effectively extract water information.
Water information extraction is an important part of water resources management. There have been many researches on water information extraction from remote sensing images in the past two decades. There are a variety of land-based water extraction methods based on satellite remote sensing. OTSU algorithm using water index is a relatively common one,but it is difficult to select the threshold and the extraction accuracy is insufficient. For this,we introduce the improved SLIC and OTSU algorithm into water information extraction.Considering the importance of water index for water information extraction,we combine SLIC,a super-pixel generation algorithm,and NDWI,the normalized water index,to generate each super-pixel,and the mean value of water index within the super-pixel represents the water index of this super-pixel. Then based on super-pixel,OTSU algorithm is used to conduct the threshold segmentation of the water index in the super-pixel. Experiment shows that the proposed algorithm not only has a high extraction accuracy,but also a fast extraction speed,which can effectively extract water information.
引文
[1]胡晓东,骆剑承,夏列钢,等.图谱迭代反馈的自适应水体信息提取方法[J].测绘学报,2011,40(5):544-550.
[2] FRAZIER P S,PAGE K J. Water body detection and delineation w ith Landsat TM data[J]. Photogrammetric Engineering and Remote Sensing,2000,66(12):1461-1468.
[3]刘建波,戴昌达. TM图象在大型水库库情监测管理中的应用[J].环境遥感,1996,11(1):54-58.
[4]邓劲松,王珂,邓艳华,等. SPOT-5卫星影像中水体信息自动提取的一种有效方法[J].上海交通大学学报:农业科学版,2005,23(2):198-201.
[5] MCFEETERS S K. The use of normalized difference water index(NDWI)in the delineation of open w ater features[J]. International Journal of Remote Sensing,1996,17(7):1425-1432.
[6]徐涵秋.利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究[J].遥感学报,2005,9(5):589-595.
[7]陈文倩,丁建丽,李艳华,等.基于国产GF-1遥感影像的水体提取方法[J].资源科学,2015,37(6):1166-1172.
[8]朱利,李云梅,赵少华,等.基于GF-1号卫星WFV数据的太湖水质遥感监测[J].国土资源遥感,2015,27(1):113-120.
[9]骆剑承,盛永伟,沈占锋,等.分步迭代的多光谱遥感水体信息高精度自动提取[J].遥感学报,2009,13(4):610-615.
[10] ACHANTA R,SHAJI A,SMITH K,et al. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Softw are Engineering,2012,34(11):2274-2282.
[11]胡蕾,侯鹏洋.一种基于光谱与纹理特征的多光谱遥感图像地物分类方法[J].中国科技论文,2015,10(2):197-200.
[12] OTSU N. A threshold selection method from gray-level histograms[J]. Automatica,1975,285-296:23-27.
[13]韩建涛,姜卫东,陈曾平.基于OTSU准则及图像熵的局部递归分割算法研究[J].红外技术,2004,26(6):89-92.
[14]高永霞,朱广伟,贺冉冉,等.天目湖水质演变及富营养化状况研究[J].环境科学,2009,30(3):673-679.
[15]朱广伟,赵林林,陈伟民,等.低水位运行对天目湖水库水质与生态的影响[J].生态与农村环境学报,2011,27(4):87-94.
[2] FRAZIER P S,PAGE K J. Water body detection and delineation w ith Landsat TM data[J]. Photogrammetric Engineering and Remote Sensing,2000,66(12):1461-1468.
[3]刘建波,戴昌达. TM图象在大型水库库情监测管理中的应用[J].环境遥感,1996,11(1):54-58.
[4]邓劲松,王珂,邓艳华,等. SPOT-5卫星影像中水体信息自动提取的一种有效方法[J].上海交通大学学报:农业科学版,2005,23(2):198-201.
[5] MCFEETERS S K. The use of normalized difference water index(NDWI)in the delineation of open w ater features[J]. International Journal of Remote Sensing,1996,17(7):1425-1432.
[6]徐涵秋.利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究[J].遥感学报,2005,9(5):589-595.
[7]陈文倩,丁建丽,李艳华,等.基于国产GF-1遥感影像的水体提取方法[J].资源科学,2015,37(6):1166-1172.
[8]朱利,李云梅,赵少华,等.基于GF-1号卫星WFV数据的太湖水质遥感监测[J].国土资源遥感,2015,27(1):113-120.
[9]骆剑承,盛永伟,沈占锋,等.分步迭代的多光谱遥感水体信息高精度自动提取[J].遥感学报,2009,13(4):610-615.
[10] ACHANTA R,SHAJI A,SMITH K,et al. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Softw are Engineering,2012,34(11):2274-2282.
[11]胡蕾,侯鹏洋.一种基于光谱与纹理特征的多光谱遥感图像地物分类方法[J].中国科技论文,2015,10(2):197-200.
[12] OTSU N. A threshold selection method from gray-level histograms[J]. Automatica,1975,285-296:23-27.
[13]韩建涛,姜卫东,陈曾平.基于OTSU准则及图像熵的局部递归分割算法研究[J].红外技术,2004,26(6):89-92.
[14]高永霞,朱广伟,贺冉冉,等.天目湖水质演变及富营养化状况研究[J].环境科学,2009,30(3):673-679.
[15]朱广伟,赵林林,陈伟民,等.低水位运行对天目湖水库水质与生态的影响[J].生态与农村环境学报,2011,27(4):87-94.