Please wait a minute...
img

官方微信

遥感技术与应用  2022, Vol. 37 Issue (5): 1217-1226    DOI: 10.11873/j.issn.1004-0323.2022.5.1217
数据与图像处理     
基于不同分辨率遥感影像自动提取切沟的精度分析和转换模型
张琪1(),张光辉2,张岩1(),王佳希1,余双武1
1.山西吉县森林生态系统国家野外科学观测研究站,北京林业大学,北京 100083
2.北京师范大学 地理科学学部,北京 100875
Accuracy Analysis and Conversion Model of Gully Automatic Extraction based on Remote Sensing Images with Different Resolutions
Qi Zhang1(),Guanhui Zhang2,Yan Zhang1(),Jiaxi Wang1,Shuangwu Yu1
1.Jixian Forest Ecosystem Studies,National Observation and Research Station,Beijing Forestry University,Beijing 100083,China
2.School of Geography,Beijing Normal University,Beijing 100875,China
 全文: PDF(4073 KB)   HTML
摘要:

黄土高原地貌类型独特而复杂,切沟侵蚀是塑造该区地貌的主要动力之一。研究不同分辨率遥感影像提取切沟的适用性和自动提取方法,可为切沟侵蚀遥感监测和沟蚀防治等提供有效手段。以黄土高原南部山西吉县残塬沟壑区为研究区,使用面向对象分析方法和随机森林分类算法分别从0.5 m Google影像、2 m GF-1融合影像和8 m GF-1多光谱影像中自动提取切沟,分析提取精度,并构建转换模型,提高低分辨率遥感影像提取的切沟沟长、面积参数的精度。结果表明:①依据特征类别,特征变量对于切沟识别的重要性排序如下:光谱特征>纹理特征>几何特征。②0.5 m和2 m分辨率影像切沟分类精度较高,生产者精度和用户精度均达90%以上,8 m GF-1影像切沟分类的生产者精度和用户精度为85%左右。③0.5 m和2 m分辨率影像提取的切沟沟长和沟宽的百分误差分别为5%和13%左右;8 m分辨率影像提取的切沟沟长、面积和沟宽的平均百分误差为18.82%、27.62%和18.93%。④基于0.5 m分辨率Google影像提取的切沟形态特征参数,建立8 m分辨率GF-1影像提取的切沟沟长转换模型(L=1.22L'-0.28)和面积转换模型(A=1.44A'+31.56),转换结果具有较高的精度。

关键词: 切沟侵蚀自动提取切沟形态参数随机森林黄土高原    
Abstract:

Gully erosion is the major driver of land degradation and the unique landforms on the Loess Plateau. It is of practical significance to assess the applicability of extracting gully from satellite images with different resolutions and explore automatic gully extraction method. Google image (0.5 m resolution) and GF-1 images (2 m and 8 m resolution) were used to extract gullies automatically with object-based image analysis and random forest in Zhongduo tableland located in the southeastern Loess Plateau. Gully morphological parameters of 30 gullies extracted from three satellite images were compared to those from UAV data (0.14 m resolution). The results were as follows: (1) The importance of image feature variables used for gully extraction is sorted as follows: spectral feature > texture feature > geometric feature. (2) The user accuracy and producer accuracy of gully extraction based on 0.5 m and 2 m resolution images were higher than 90%, while the user accuracy and producer accuracy reduced to 85% when 8 m resolution image was used. (3) The errors of gully length and width extracted from 0.5 m and 2 m resolution images were about 5% and 13%. The average error of extracted gully length, area and width from 8 m resolution image were 18.82%, 27.62% and 18.93%, respectively. (4) A model was put forward for improving the accuracy of gully length and gully area extracted from GF-1 image with 8 m resolution, based on the gully parameters extracted from 0.5 m resolution image, i.e., L=1.22L'- 0.28, R2=0.896 and A=1.44A'+ 31.56, R2=0.916.

Key words: Gully erosion    Automatic extraction    Gully morphological parameters    Random forest    Loess Plateau
收稿日期: 2021-07-21 出版日期: 2022-12-13
ZTFLH:  K903  
基金资助: 国家自然科学基金重点项目“黄土高原植被恢复影响切沟侵蚀的动力机制与模拟”(42130701);国家自然科学基金项目“黄土高塬沟壑区沟头溯源与沟谷不同微地貌侵蚀过程和机制”(42177309)
通讯作者: 张岩     E-mail: 810120269@qq.com;zhangyan9@bjfu.edu.cn
作者简介: 张 琪(1997—),女,河北石家庄人,硕士研究生,主要从事3S技术集成开发与应用研究。E?mail: 810120269@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
张琪
张光辉
张岩
王佳希
余双武

引用本文:

张琪,张光辉,张岩,王佳希,余双武. 基于不同分辨率遥感影像自动提取切沟的精度分析和转换模型[J]. 遥感技术与应用, 2022, 37(5): 1217-1226.

Qi Zhang,Guanhui Zhang,Yan Zhang,Jiaxi Wang,Shuangwu Yu. Accuracy Analysis and Conversion Model of Gully Automatic Extraction based on Remote Sensing Images with Different Resolutions. Remote Sensing Technology and Application, 2022, 37(5): 1217-1226.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2022.5.1217        http://www.rsta.ac.cn/CN/Y2022/V37/I5/1217

图1  研究区位置及实地照片
特征类型特征名称简称特征数量
光谱特征均值Mean R, Mean G, Mean B, Mean NIR4
标准差Sd R, Sd G, Sd B, Sd NIR4
波段最大差异MaxDiff1
总体亮度值Brightness1
波段比率ratio_GB, ratio_RB, ratio_RG,ratio_RN4
归一化植被指数NDVI1
纹理特征均值GL_M R, GL_M G, GL_M B, GL_M NIR4
标准差GL_S R, GL_S G, GL_S B, GL_S NIR4
GL_E R, GL_E G, GL_E B, GL_E NIR4
同质度GL_H R, GL_H G, GL_H B, GL_H NIR4
对比度GL_Con R, GL_Con G, GL_Con B, GL_Con NIR4
非相似性GL_D R, GL_D G, GL_D B, GL_D NIR4
角二阶矩GL_A R, GL_A G, GL_A B, GL_A NIR4
相关性GL_Cor R, GL_Cor G, GL_Cor B, GL_Cor NIR4
几何特征面积Area1
长宽比LW1
长度Length1
形状指数SI1
不对称性Asymmetry1
圆度Roundness1
紧致度Compactness1
矩形拟合RF1
表 1  用于提取切沟的影像特征列表
影像数据尺度形状紧致度过分割欠分割欧氏距离
Google3900.20.70.1250.1020.114
3730.30.80.1190.0800.102
4390.40.80.1490.0810.120
3830.30.90.1160.0970.107
3980.40.70.1090.1080.109

GF-1

(2 m)

1060.10.90.1660.1180.144
870.20.80.1430.1350.139
840.30.70.1500.1150.134
950.20.90.1570.1200.140
860.30.80.1730.1260.151
940.30.90.1620.1130.140
810.40.80.1520.1230.138

GF-1

(8 m)

210.20.50.2120.1890.201
200.20.60.1820.1630.173
240.30.60.1850.1700.178
170.30.70.1960.1750.186
210.40.80.2030.1820.193
表2  Google影像和GF-1影像分割精度评价
图2  不同分辨率影像局部方差、方差变化率变化曲线
图3  不同分辨率影像特征数量与分类总体精度的关系
图4  不同分辨率影像特征重要性分布
影像数据沟谷地农用地建设用地未利用地总体精度/%

Kappa

系数

生产者精度/%

用户

精度/%

生产者精度/%

用户

精度/%

生产者

精度/%

用户

精度/%

生产者精度/%

用户

精度/%

Google(0.5 m)95.5894.6793.5572.5084.5294.6795.5898.1891.460.87
GF-1(2 m)90.493.5789.0470.2778.2675.0089.0479.2787.010.77
GF-1(8 m)82.9085.1485.4558.7577.0282.6082.9085.1481.560.72
表3  Google影像和GF-1影像分类精度评价
图5  Google影像和GF-1影像中提取的沟缘线
参数影像数据均值/m中值/m最大值/m最小值/m平均绝对误差平均百分误差/%
沟长L/m无人机126.18109.05317.1422.60//
Google124.44106.73317.4120.234.285.35
GF-1(2 m)119.7999.49333.0420.8210.829.36
GF-1(8 m)102.6188.24289.3219.4223.9618.82
面积A/m2无人机6 159.905 126.7919 576.97436.91//
Google6 160.835 139.5519 127.49490.98368.616.81
GF-1(2 m)5 613.124 822.9114 876.00336.00799.0012.25
GF-1(8 m)4 320.013 024.0513 964.34358.141 853.2127.62
沟宽W/m无人机42.8644.0772.8619.33//
Google43.0144.5167.8820.212.265.22
GF-1(2 m)42.5943.2767.7216.135.3313.40
GF-1(8 m)37.4237.3262.9916.808.6818.93
表4  切沟形态参数统计结果
图6  Google影像和GF-1影像提取切沟形态参数百分误差箱图
图7  切沟沟长和面积转换模型建立与验证
1 Li Y, Poesen J, Yang J C, et al. Evaluating gully erosion using 137Cs and 210Pb /137Cs ratio in a catchment[J]. Soil & Tillage Research,2003,69(1/2):107-115. DOI:10.1016/S0167-1987(02)00132-0 .
doi: 10.1016/S0167-1987(02)00132-0
2 Zhang Guanghui. Advances and prospects for gully erosion researches[J]. Journal of Soil and Water Conservation, 2020, 34(5):1-13.
2 张光辉.切沟侵蚀研究进展与展望[J].水土保持学报,2020,34(5):1-13.
3 Singh B M. Interpretation of satellite imagery for delineation of ravines[J]. Journal of the Indian Society of Photo-Interpretation, 1977, 5(1):31-34. DOI: 10.1007/BF03025387 .
doi: 10.1007/BF03025387
4 Karami A, Khoorani A, Noohegar A, et al. Gully erosion mapping using object-based and pixel-based image classification methods[J]. Environmental & Engineering Geoscience, 2015,21(2):101-110. DOI:10.2113/gseegeosci.21.2.101 .
doi: 10.2113/gseegeosci.21.2.101
5 Eustace A, Pringle M, Witte C. Give me the dirt: detection of gully extent and volume using high-resolution LiDAR[C]∥ Jones S,Reinke K. Innovations in Remote sensing and Photogrammetry Berlin Heidelberg:Springer,2009:255-269. DOI:10.1007/978-3-540-93962-7_20 .
doi: 10.1007/978-3-540-93962-7_20
6 Shruthi R, Kerle N, Jetten V. Object-based gully feature extraction using high spatial resolution imagery[J]. Geomorphology,2011,134(3/4):260-268.DOI:10.1016/j.geomorph. 2011.07.003 .
doi: 10.1016/j.geomorph. 2011.07.003
7 Zhou Peicheng, Cheng Gong, Yao Xiwen,et al. Machine learning paradigms in high-resolution remote sensing image interpretation[J].National Remote Sensing Bulletin,2021,25(1):182-197.
7 周培诚,程塨,姚西文,等.高分辨率遥感影像解译中的机器学习范式[J].遥感学报,2021,25(1):182-197.
8 Tang Tingyuan, Fu Bolin, He Suyun, et al. Identification of typical land features in the Lijiang River basin with fusion optics and radar[J]. Remote Sensing Technology and Application,2020,35(2):448-457.
8 唐廷元,付波霖,何素云,等.基于GF-1和Sentinel-1A的漓江流域典型地物信息提取[J].遥感技术与应用,2020,35(2):448-457.
9 Shruthi R, Kerle N, Jetten V,et al.Quantifying temporal chan-ges in gully erosion areas with object oriented analysis[J]. Catena,2015,128:262-267. DOI:10.1016/j.catena.2014. 01.010 .
doi: 10.1016/j.catena.2014. 01.010
10 Shahabi H, Jarihani B, Piralilou S T, et al. A semi-automated object-based gully networks detection using different machine learning models: A case study of bowen catchment, Queensland,Australia[J]. Sensors,2019,19(22):4893. DOI: 10.3390/s19224893 .
doi: 10.3390/s19224893
11 Garosi Y, Sheklabadi M, Pourghasemi H R, et al. Comparison of differences in resolution and sources of controlling factors for gully erosion susceptibility mapping[J]. Geoderma, 2018,330:65-78. DOI: 10.1016/j.geoderma.2018.05.027 .
doi: 10.1016/j.geoderma.2018.05.027
12 Phinzi K, Holb I, Szabo S. Mapping permanent gullies in an agricultural area using satellite images: Efficacy of machine learning algorithms[J]. Agronomy,2021,11:333. DOI:10. 3390/agronomy11020333 .
doi: 10. 3390/agronomy11020333
13 Pu Luoman, Zhang Shuwen, Wang Ranghu, et al. Analysis of erosion gully information extraction based on multi-resource remote sensing images[J]. Geography and Geo-Information Science, 2016, 32(1):90-94.
13 蒲罗曼,张树文,王让虎,等.多源遥感影像的侵蚀沟信息提取分析[J].地理与地理信息科学,2016,32(1):90-94.
14 Yao Yuan, Chen Xi, Qian Jing. A review on the methodology of scale issues in quantitative remote sensing[J]. Scientia Geographica Sinica, 2019,39(3):367-376.
14 姚远,陈曦,钱静.定量遥感尺度转换方法研究进展[J].地理科学,2019,39(3):367-376.
15 Liu Yan, Wang Jindi, Zhou Hongmin, et al. LAI measuring data processing, analysis and spatial scaling in the middle reaches of Heihe experimental research region[J].Remote Sen-sing Technology and Application, 2010,25(6):805-813.
15 刘艳,王锦地,周红敏,等.黑河中游试验区不同分辨率LAI数据处理、分析和尺度转换[J].遥感技术与应用,2010,25(6):805-813.
16 Ji Menghao, Tang Bohui, Li Zhaoliang. Review of solar-induced chlorophyll fluorescence retrieval methods from satellite data[J]. Remote Sensing Technology and Application, 2019,34(3):455-466.
16 纪梦豪,唐伯惠,李召良.太阳诱导叶绿素荧光的卫星遥感反演方法研究进展[J].遥感技术与应用,2019,34(3):455-466.
17 Zhu Xinming, Song Xiaoning, Leng Pei, et al. Spatial downscaling of land surface temperature with the multi-scale geographically weighted regression[J]. National Remote Sensing Bulletin, 2021,25(8):1749-1766.
17 祝新明,宋小宁,冷佩,等.多尺度地理加权回归的地表温度降尺度研究[J].遥感学报,2021,25(8):1749-1766.
18 Shi Quanzheng. Jixian County Chronicle[M]. Beijing: China Science and Technology Press, 1992:50-58.
18 石全正.吉县志[M].北京:中国科学技术出版社,1992:50-58.
19 Mei X, Ma L, Zhu Q, et al. The variability in soil water storage on the loess hillslopes in China and its estimation[J]. Catena,2019,172:807-818. DOI:10.1016/j.catena.2018.09.045 .
doi: 10.1016/j.catena.2018.09.045
20 Luo Laixing.A tentative classification of landforms in the Loess Plateau[J].Acta Geographica Sinica,1956,22(3):201-222.
20 罗来兴.划分晋西、陕北、陇东黄土区域沟间地与沟谷的地貌类型[J].地理学报,1956,22(3):201-222.
21 Zheng Fenli, Xu Ximeng, Qin Chao. A review of gully erosion process research[J]. Transactions of the Chinese Society for Agricultural Machinery,2016,47(8):48-59,116.
21 郑粉莉,徐锡蒙,覃超.沟蚀过程研究进展[J].农业机械学报,2016,47(8):48-59,116.
22 Dragut L, Csillik O, Eisank C, et al. Automated parameterisation for multi-scale image segmentation on multiple layers[J].ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 88: 119-127. DOI: 10.1016/j.isprsjprs.2013.11.018 .
doi: 10.1016/j.isprsjprs.2013.11.018
23 Clinton N, Holt A, Scarborough J, et al. Accuracy assessment measures for object-based image segmentation goodness[J]. Photogrammetric Engineering and Remote Sensing, 2010,76(3):289-299. DOI: 10.14358/PERS.76.3.289 .
doi: 10.14358/PERS.76.3.289
24 Liu K, Ding H, Tang G, et al. An object-based approach for two-level gully feature mapping using high-resolution DEM and imagery: A case study on hilly Loess Plateau region, China[J]. Chinese Geographical Science, 2017, 27(3): 415-430. DOI: 10.1007/s11769-017-0874-x .
doi: 10.1007/s11769-017-0874-x
25 Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32. DOI: 10.1023/A:1010933404324 .
doi: 10.1023/A:1010933404324
26 Ghorbanzadeh O, Blaschke T, Gholamnia K, et al. Evaluation of different machine learning methods and deep-learning convolutional neural networks for landslide detection[J]. Remote Sensing, 2019, 11(2):196. DOI:10.3390/rs11020196 .
doi: 10.3390/rs11020196
27 Li Zhen, Zhang Yan, Yang Song, et al. Error assessment of extracting morphological parameters of bank gullies by manual visual interpretation based on QuickBird imagery[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(20):179-186.
27 李镇,张岩,杨松,等.QuickBird影像目视解译法提取切沟形态参数的精度分析[J].农业工程学报,2014,30(20):179-186.
28 Liu K, Hu D, Tang G, et al. Detection of catchment-scale gully-affected areas using Unmanned Aerial Vehicle (UAV) on the Chinese Loess Plateau[J]. International Journal of Geo-Information, 2016, 5(12):238. DOI: 10.3390/ijgi5120238 .
doi: 10.3390/ijgi5120238
29 Wu Hongyan, Zheng Fenli, Xu Ximeng, et al. Transformation of gully morphological characteristic indicators extracted from DEMs of different resolution[J]. Journal of Soil and Water Conservation,2016,30(6):147-152,161.
29 吴红艳,郑粉莉,徐锡蒙,等.不同分辨率DEM提取切沟形态特征参数的转化研究[J].水土保持学报,2016,30(6):147-152,161.
[1] 马超,黄华国,田昕,刘炳杰,温坤剑,王鹏杰. 基于背包激光雷达的单木材积估测方法研究[J]. 遥感技术与应用, 2022, 37(5): 1071-1083.
[2] 高帅,侯学会,汪云,王倩,陈悦,邢瑞,王晶. 基于机器学习和大数据平台的陆地生态系统碳收支遥感监测[J]. 遥感技术与应用, 2022, 37(5): 1190-1197.
[3] 丁书培,李蒙蒙,汪小钦,李琳,吴瑞姣,黄姮. 基于时序遥感数据的福州市耕地非农化特征及驱动因子分析[J]. 遥感技术与应用, 2022, 37(3): 550-563.
[4] 李媛,周小成,陈芸芝,王锋克. 结合Landsat时序遥感干扰检测的福建将乐县森林林龄估算与评价[J]. 遥感技术与应用, 2022, 37(3): 651-662.
[5] 王美雅,徐涵秋. 超大城市土地覆盖与热环境的随机森林回归模型研究[J]. 遥感技术与应用, 2022, 37(2): 379-388.
[6] 康新礼,张文豪,刘原萍,顾行发,余涛,张丽丽,徐桦昆. 基于随机森林的京津冀地区PM2.5遥感反演及变化分析[J]. 遥感技术与应用, 2022, 37(2): 424-435.
[7] 李翔华,黄春林,侯金亮,韩伟孝,冯娅娅,陈彦四,王静. 结合遥感和统计数据的家畜分布网格化方法研究[J]. 遥感技术与应用, 2022, 37(1): 262-271.
[8] 姜玉峰,齐建国,陈博伟,闫敏,黄龙吉,张丽. 基于无人机高光谱影像和机器学习的红树林树种精细分类[J]. 遥感技术与应用, 2021, 36(6): 1416-1424.
[9] 常潇月,荆林海,林沂. 基于机载LiDAR数据和Landsat 8影像的芬兰北部泰加林—苔原过渡带提取[J]. 遥感技术与应用, 2021, 36(6): 1259-1271.
[10] 魏雪馨,刘洋,闵庆文,刘荣高,张清洋,叶晓星,刘蓓蓓. 基于地物光谱季节曲线特征的毛竹林分布提取[J]. 遥感技术与应用, 2021, 36(5): 1178-1188.
[11] 姚杰鹏,杨磊库,陈探,宋春桥. 基于Sentinel-1,2和Landsat 8时序影像的鄱阳湖湿地连续变化监测研究[J]. 遥感技术与应用, 2021, 36(4): 760-776.
[12] 李晓彤,覃先林,刘倩,赵俊鹏,王崇阳. 基于AISA Eagle Ⅱ机载高光谱数据的森林可燃物类型识别方法[J]. 遥感技术与应用, 2021, 36(3): 544-551.
[13] 盛夏,石玉立,丁海勇. 青藏高原GPM降水数据空间降尺度研究[J]. 遥感技术与应用, 2021, 36(3): 571-580.
[14] 陈彦四,黄春林,侯金亮,韩伟孝,冯娅娅,李翔华,王静. 基于多时相Sentinel-2影像的黑河中游玉米种植面积提取研究[J]. 遥感技术与应用, 2021, 36(2): 324-331.
[15] 杨福芹,冯海宽,李振海,潘洁晨,谢瑞. 苹果叶片氮含量高光谱反演方法对比[J]. 遥感技术与应用, 2021, 36(2): 353-361.