基于TM遥感影像溢油识别方法的比较研究
|
摘要
为了及时掌握溢油信息,实现对溢油区域的高精度监测,采用平行管道法、最大似然法和支持向量机法3种监督分类方法识别两幅TM影像的溢油区域,分析各种分类法在溢油识别领域的差异和其适用范围,同时利用绿波段和反射红外波段的有效组合对溢油区域的范围加以确定。结果表明:支持向量机分类法在两景TM遥感影像溢油识别上的总体精度和Kappa系数均为最高,过分割误差与欠分割误差较低,且时间复杂度较为合理,能够对大面积溢油实现较好的监测。
In order to timely grasp the oil spill information and achieve high-precision monitoring of the oil spill area,the supervised classification method( parallel hexahedron method,maximum likelihood method and support vector machine method)is used to identify the oil spill areas of the two TM images,and analyzes various classifications. The difference between the method in the field of oil spill identification and its scope of application,while using the effective combination of the green band and the reflected infrared band to determine the range of the oil spill area. The results show that the overall accuracy and Kappa coefficient of the SVM classification method are the highest in the two-view TM image oil spill recognition. Over-segmentation error and under-segmentation error are low,and the time complexity is reasonable,which can achieve better monitoring of large-area oil spill.
In order to timely grasp the oil spill information and achieve high-precision monitoring of the oil spill area,the supervised classification method( parallel hexahedron method,maximum likelihood method and support vector machine method)is used to identify the oil spill areas of the two TM images,and analyzes various classifications. The difference between the method in the field of oil spill identification and its scope of application,while using the effective combination of the green band and the reflected infrared band to determine the range of the oil spill area. The results show that the overall accuracy and Kappa coefficient of the SVM classification method are the highest in the two-view TM image oil spill recognition. Over-segmentation error and under-segmentation error are low,and the time complexity is reasonable,which can achieve better monitoring of large-area oil spill.
引文
[1] YIN L,ZHANG M,ZHANG Y,et al. The longterm prediction of the oil-contaminated water from the sanchi collision in the east China sea[J]. Acta Oceanologica Sinica,2018,37(3):69-72.
[2]张彦敏,徐卓,旭锋.海上溢油合成孔径雷达探测研究[J].中国海洋大学学报(自然科学版),2017,47(2):106-115.
[3]孙元芳,周斌,盖颖颖,等.遥感监测墨西哥湾溢油目标识别算法[J].测绘科学,2015,40(3):63-67.
[4] HOU Y,GUO H,WANG N,et al. Research on the classification for MODIS images of oil slick on the sea[C]∥2012 IEEE 11th International Conference on Signal Processing(ICSP). Beijing:IEEE,2012:966-971.
[5]范剑超,王涛.基于非负矩阵分解和支持向量机的墨西哥湾溢油HJ-1星遥感图像分类[J].海洋环境科学,2015,34(3):441-446.
[6] LIU S,CHI M,ZOU Y,et al. Oil spill detection via multitemporal optical remote sensing images:a change detection perspective[J]. IEEE Geoscience&Remote Sensing Letters,2017,14(3):324-328.
[7] LEE M S,PARK K A,LEE H R,et al. Detection and dispersion of thick and film-like oil spills in a coastal bay using satellite optical images[J]. IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2016,9(11):5139-5150.
[8] LANEVE G,LUCIANI R. Developing a satellite optical sensor based automatic system for detecting and monitoring oil spills[C]∥2015 IEEE 15th International Conference on Environment and Electrical Engineering(EEEIC). Rome:IEEE,2015:1653-1658.
[9]邵琦,王思懿,杨春熙,等. Landsat 5 TM、Landsat 7ETM+及Landsat 8 OLI交叉定标研究[J].产业与科技论坛,2017,16(17):54-57.
[10]温素馨,韦玉春,汪美会. TM遥感图像FLAASH大气校正异常值的改正[J].测绘科学,2017,42(7):165-171.
[11] ADAMO M,DE C G,DE P V,et al. Detection and tracking of oil slicks on sun-glittered visible and near infrared satellite imagery[J]. International Journal of Remote Sensing,2009,30(24):6403-6427.
[12]李阳,常霞,纪峰.图像增强方法研究新进展[J].传感器与微系统,2015,34(12):9-12.
[13]付仲良,陈静.多时相Landsat数据动态监测城市土地利用变化:以上海市为例[J].地理信息世界,2016,23(6):26-30.
[14]王增林,朱大明.基于遥感影像的最大似然分类算法的探讨[J].河南科学,2010,28(11):1458-1461.
[15]王立国,赵亮,刘丹凤. SVM在高光谱图像处理中的应用综述[J].哈尔滨工程大学学报,2018,39(6):973-978.
[16]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.
[17] KEERTHI S S,LIN C J. Asymptotic behaviors of support vector machines with gaussian kernel[J].Neural Computation,2003,15(7):1667-1689.
[18]王臣,刘朋,赵朝方,等.高分一号在青岛输油管道海面溢油监测中的应用[J].海洋湖沼通报,2014(4):174-179.
[19]陈辉,赵朝方. MODIS多光谱信息在海上溢油检测
[2]张彦敏,徐卓,旭锋.海上溢油合成孔径雷达探测研究[J].中国海洋大学学报(自然科学版),2017,47(2):106-115.
[3]孙元芳,周斌,盖颖颖,等.遥感监测墨西哥湾溢油目标识别算法[J].测绘科学,2015,40(3):63-67.
[4] HOU Y,GUO H,WANG N,et al. Research on the classification for MODIS images of oil slick on the sea[C]∥2012 IEEE 11th International Conference on Signal Processing(ICSP). Beijing:IEEE,2012:966-971.
[5]范剑超,王涛.基于非负矩阵分解和支持向量机的墨西哥湾溢油HJ-1星遥感图像分类[J].海洋环境科学,2015,34(3):441-446.
[6] LIU S,CHI M,ZOU Y,et al. Oil spill detection via multitemporal optical remote sensing images:a change detection perspective[J]. IEEE Geoscience&Remote Sensing Letters,2017,14(3):324-328.
[7] LEE M S,PARK K A,LEE H R,et al. Detection and dispersion of thick and film-like oil spills in a coastal bay using satellite optical images[J]. IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2016,9(11):5139-5150.
[8] LANEVE G,LUCIANI R. Developing a satellite optical sensor based automatic system for detecting and monitoring oil spills[C]∥2015 IEEE 15th International Conference on Environment and Electrical Engineering(EEEIC). Rome:IEEE,2015:1653-1658.
[9]邵琦,王思懿,杨春熙,等. Landsat 5 TM、Landsat 7ETM+及Landsat 8 OLI交叉定标研究[J].产业与科技论坛,2017,16(17):54-57.
[10]温素馨,韦玉春,汪美会. TM遥感图像FLAASH大气校正异常值的改正[J].测绘科学,2017,42(7):165-171.
[11] ADAMO M,DE C G,DE P V,et al. Detection and tracking of oil slicks on sun-glittered visible and near infrared satellite imagery[J]. International Journal of Remote Sensing,2009,30(24):6403-6427.
[12]李阳,常霞,纪峰.图像增强方法研究新进展[J].传感器与微系统,2015,34(12):9-12.
[13]付仲良,陈静.多时相Landsat数据动态监测城市土地利用变化:以上海市为例[J].地理信息世界,2016,23(6):26-30.
[14]王增林,朱大明.基于遥感影像的最大似然分类算法的探讨[J].河南科学,2010,28(11):1458-1461.
[15]王立国,赵亮,刘丹凤. SVM在高光谱图像处理中的应用综述[J].哈尔滨工程大学学报,2018,39(6):973-978.
[16]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.
[17] KEERTHI S S,LIN C J. Asymptotic behaviors of support vector machines with gaussian kernel[J].Neural Computation,2003,15(7):1667-1689.
[18]王臣,刘朋,赵朝方,等.高分一号在青岛输油管道海面溢油监测中的应用[J].海洋湖沼通报,2014(4):174-179.
[19]陈辉,赵朝方. MODIS多光谱信息在海上溢油检测