[1]罗凤宇,高艺非,谢勇,等.基于随机森林算法与多时相Sentinel-2影像数据的茶树种植区信息提取[J].江苏农业学报,2024,(09):1671-1680.[doi:doi:10.3969/j.issn.1000-4440.2024.09.011]
　LUO Fengyu,GAO Yifei,XIE Yong,et al.Extraction of tea plantation area information based on random forest algorithm and multi-temporal Sentinel-2 image data[J].,2024,(09):1671-1680.[doi:doi:10.3969/j.issn.1000-4440.2024.09.011]
点击复制

基于随机森林算法与多时相Sentinel-2影像数据的茶树种植区信息提取()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2024年09期
页码:
1671-1680
栏目:
农业信息工程
出版日期:
2024-09-30

文章信息/Info

Title:
Extraction of tea plantation area information based on random forest algorithm and multi-temporal Sentinel-2 image data
作者:
罗凤宇1; 高艺非2; 谢勇1; 邹旭辉1; 邵雯1; 张世雨1
(1.南京信息工程大学地理科学学院,江苏南京210044；2.江苏师范大学地理测绘与城乡规划学院,江苏徐州221116)
Author(s):
LUO Fengyu1; GAO Yifei2; XIE Yong1; ZOU Xuhui1; SHAO Wen1; ZHANG Shiyu1
(1.School of Geographical Sciences, Nanjing University of Information Science & Technology, Nanjing 210044, China;2.School of Geography, Geomatics and Planning, Jiangsu Normal University, Xuzhou 221116, China)
关键词:
茶树种植区; 随机森林; 多时相特征; 面积监测
Keywords:
tea plantation area; random forest; multi-temporal features; area monitoring
分类号:
S127
DOI:
doi:10.3969/j.issn.1000-4440.2024.09.011
文献标志码:
A
摘要:
茶树是中国重要的木本经济作物,及时准确地获取茶树种植区面积及空间分布对区域农业经济发展具有重要意义。本研究以安徽省郎溪县为研究区,首先分析茶树、小麦和红叶石楠时序光谱特征,其次基于3个时相的Sentinel-2影像数据提取光谱特征、水体指数及植被指数特征、红边指数特征、纹理特征组成多时相特征变量数据集,并设置6种特征变量组合方案,利用随机森林算法进行茶树种植区信息提取精度的比较,筛选得到适宜的特征变量组合方案,最后基于适宜的特征变量组合方案进行郎溪茶树种植区信息的提取。结果表明,在光谱特征变量的基础上,分别融合水体指数及植被指数特征、红边指数特征和纹理特征变量均能有效提高茶树种植区信息的提取精度,其中,红边指数特征对茶树种植区信息提取精度的提高效果最好,其次是水体指数及植被指数特征。基于随机森林-平均精确度下降算法(RF-MDA)优选后的特征变量组合的分类效果最佳,总体分类精度达94.95%,Kappa系数为0.934 8,说明特征变量优选能有效地保留重要的地物识别特征变量,避免冗余信息对分类结果的影响。综上,基于随机森林算法和茶树多时相Sentinel-2影像数据能实现郎溪县茶树种植区信息的高精度提取。
Abstract:
Tea plant is an important woody cash crop in China, and timely and accurate acquisition of the planting area and spatial distribution of tea plants is of great significance to regional agricultural economic development. In this study, Langxi County in Anhui province was taken as the study area. Firstly, the temporal spectral characteristics of tea trees, wheat and Photinia fraseri were analyzed. Secondly, based on the Sentinel-2 image data of three phases, spectral features, water index and vegetation index features, red edge index features and texture features were extracted to form a multi-temporal feature variable data set, and six feature variable combination schemes were set up. The random forest algorithm was used to compare the extraction accuracy of tea tree planting areas, and the suitable combination scheme of characteristic variables was selected. Finally, the tea tree planting area in Langxi County was extracted based on the suitable combination scheme of characteristic variables. The results showed that on the basis of spectral feature variables, the fusion of water index and vegetation index features, red edge index features and texture feature variables could effectively improve the extraction accuracy of tea planting area information. Among them, red edge index features had the best effect on improving the extraction accuracy of tea planting areas, followed by water index and vegetation index features. The classification effect of the feature variable combination based on the random forest-average accuracy reduction algorithm (RF-MDA) was the best, the overall classification accuracy was 94.95%, and the Kappa coefficient was 0.934 8, indicating that the feature variable optimization could effectively retain the important feature recognition feature variables and avoid the influence of redundant information on the classification results. In summary, based on the random forest algorithm and the multi-temporal Sentinel-2 image data of tea trees, the high-precision extraction of tea planting area information in Langxi County can be realized.

参考文献/References:

[1]肖智,黄贤金,孟浩,等. 2009-2014年中国茶叶生产空间演变格局及变化特征[J]. 地理研究,2017,36(1):109-120.
[2]梁继,郑镇炜,夏诗婷,等. 高分六号红边特征的农作物识别与评估[J]. 遥感学报,2020,24(10):1168-1179.
[3]巫明焱,董光,税丽,等. 基于Landsat 8影像的济宁市春季主要作物种植面积变化监测[J]. 江苏农业学报,2018,34(3):559-569.
[4]KHATAMI R, MOUNTRAKIS G, STEHMAN S V. A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes:general guidelines for practitioners and future research[J]. Remote Sensing of Environment,2016,177:89-100.
[5]徐乔,张霄,余绍淮,等. 综合多特征的极化SAR图像随机森林分类算法[J]. 遥感学报,2019,23(4):685-694.
[6]王娜,李强子,杜鑫,等. 单变量特征选择的苏北地区主要农作物遥感识别[J]. 遥感学报,2017,21(4):519-530.
[7]张猛,曾永年,朱永森. 面向对象方法的时间序列MODIS数据湿地信息提取——以洞庭湖流域为例[J]. 遥感学报,2017,21(3):479-492.
[8]柳文杰,曾永年,张猛. 融合时间序列环境卫星数据与物候特征的水稻种植区提取[J]. 遥感学报,2018,22(3):381-391.
[9]XU W, HUANG S, WU C, et al. The pruning phenological phase-based method for extracting tea plantations by field hyperspectral data and Landsat time series imagery[J]. Geocarto International,2020,37(7):2116-2136.
[10]DAVRANCHE A,POULIN B,LEFEBVRE G. Mapping flooding regimes in Camargue wetlands using seasonal multispectral data[J]. Remote Sensing of Environment,2013,138:165-171.
[11]平跃鹏,臧淑英. 基于MODIS时间序列及物候特征的农作物分类[J]. 自然资源学报,2016,31(3):503-513.
[12]王斌,何丙辉,林娜,等. 基于随机森林特征选择的茶园遥感提取[J]. 吉林大学学报(工学版),2022,52(7):1719-1722.
[13]SHOKO C, MUTANGA O. Examining the strength of the newly-launched Sentinel 2 MSI sensor in detecting and discriminating subtle differences between C2 and C4 grass species[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2017,129:22-40.
[14]KORHONEN L, HADI, PACKALEN P, et al. Comparison of Sentinel-2 and Landsat 8 in the estimation of boreal forest canopy cover and leaf area index[J]. Remote Sensing of Environment,2017,195:259-274.
[15]郑阳,吴炳方,张淼. Sentinel-2数据的冬小麦地上干生物量估算及评价[J]. 遥感学报,2017,21(2):218-228.
[16]陈炳环. 茶树品种分类的探讨[J]. 中国茶叶,1988(2):16-18.
[17]金志凤,黄敬蜂,李波,等. 基于GIS及气候-土壤-地形因子的浙江省茶树栽培适宜性评价[J]. 农业工程学报,2017,27(3):231-236.
[18]李时睿,王治海,杨再强,等. 江南茶区茶叶生产现状和气候资源特征分析[J]. 干旱气象,2014,32(6):1007-1014.
[19]国家质量监督检验检疫总局,国家标准化管理委员会. 土地利用现状分类:GB/T 21010-2017[S]. 北京:中国标准出版社,2017.
[20]刘纪远. 中国资源环境遥感宏观调查与动态[M]. 北京:中国科学技术出版社,1996.
[21]张景华,封志明,姜鲁光. 土地利用/土地覆被分类系统研究进展[J].资源科学,2011,33(6):1195-1203.
[22]童庆禧,张兵,郑兰芬. 高光谱遥感-原理、技术与应用[M].北京:高等教育出版社,2006.
[23]HARALICK R M, SHANMUGAM K. Textural features for image classification[J]. IEEE Transactions on Systems,Man,and Cybernetics,1973,3(6):610-621.
[24]XIAO X M, BOLES S, LIU J Y, et al. Mapping paddy rice agriculture in Southern China using multi-temporal MODIS images[J]. Remote Sensing of Environment,2005,95(4):480-492.
[25]GAO B C. NDWI:a normalized difference water index for remote sensing of vegetation liquid water from space[J]. Remote Sensing of Environment,1996,58(2):257-266.
[26]ERJR H, ROCK B N. Detection of changes in leaf water content using near-and middle-infrared reflectances[J]. Remote Sensing of Environ,1989,20(1):42-54.
[27]RONDEAUX G, STEVEN M, BARET F. Optimization of soil-adjusted vegetation indices[J]. Remote Sensing of Environment,1996,55(2):95-107.
[28]BIRTH C S, MCVEY G. Measuring the color of growing turf with a reflectance spectrophotometer[J]. Agronomy Journal,1968,60:640-643.
[29]TUCKER C J. Red and photographic infrared linear combinations for monitoring vegetation[J]. Remote Sensing of Environment,1979,8(2):127-150.
[30]MIURA T, HUETE A R, YOSHIOKA H. Evaluation of sensor calibration uncertainties on vegetation indices for MODIS[J]. IEEE Transactions on Geoscience ＆ Remote Sensing,2000,38(3):1399-1409.
[31]GAMON J A, PEUELAS J, FIELD C B. A narrow-waveband spectral index that tracks diurnal changes in photosynthetic efficiency[J]. Remote Sensing of Environment,1992,41(1):25-44.
[32]GITELSON A A， VIA A, ARKEBAUER T J, et al. Remote estimation of leaf area index and green leaf biomass in maize canopies[J]. Geophysical Research Letters,2003,30(5):1248.
[33]DEVENTER V, WARD A D, GOWDA P H, et al. Using thematic mapper data to identify contrasting soil plains and tillage practices[J]. Photogrammetric Engineering & Remote Sensing,1997,62(1):87-92.
[34]GITELSON A A, MERZLYAK M N. Remote estimation of chlorophyll content in higher plant leaves[J]. International Journal of Remote Sensing,1997,18(12):2691-2697.
[35]张磊,宫兆宁,王启为,等. Sentinel-2影像多特征优选的黄河三角洲湿地信息提取[J]. 遥感学报,2019,22(2):212-226.
[36]GITELSON A A, GRITZ Y, MERZLYAK M N. Relationships between leaf chlorophyll content and spectral reflectance and algorithms for non-destructive chlorophyll assessment in higher plant leaves[J]. Journal of Plant Physiology,2002,160(2):271-282.
[37]MERZLYAK M N, GITELSON A A, CHIVKUNOVA O B, et al. Non-destructive optical detection of pigment changes during leaf senescence and fruit ripening[J]. Physiologia Plantarum,1999,106(1):125-141.
[38]FRAMPTON W J, DASH J, WATMOUGH G, et al. Evaluating the capabilities of Sentinel-2 for quantitative estimation of biophysical variables in vegetation[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2012,82:82-92.
[39]WU C Y, NIU Z, TANG Q, et al. Estimating chlorophyll content from hyperspectral vegetation indices:modeling and validation[J]. Agricultural and Forest Meteorology,2008,148(8/9):1220-1241.
[40]BREIMAN L. Random forests[J]. Machine Learning,2001,45(1):5-22.
[41]GENUER R, POGGI J M, TULEAU-MALOT C. Variable selection using random forests[J]. Pattern Recognition Letters,2010,31(14):2225-2236.
[42]CONGALTON R G. A review of assessing the accuracy of classifications of remotely sensed data[J]. Remote Sensing of Environment,1991,37(1):35-46.

相似文献/References:

[1]牛芳鹏,李新国,麦麦提吐尔逊·艾则孜,等.基于光谱指数的博斯腾湖西岸湖滨绿洲土壤有机碳含量估算模型[J].江苏农业学报,2022,38(02):414.[doi:doi:10.3969/j.issn.1000-4440.2022.02.015]
　NIU Fang-peng,LI Xin-guo,MAMATTURSUN·Eziz,et al.Estimation model of soil organic carbon content in lakeside oasis on the west coast of Bosten Lake based on spectral index[J].,2022,38(09):414.[doi:doi:10.3969/j.issn.1000-4440.2022.02.015]
[2]李长春,翟伟广,王春阳,等.基于Sentinel-1A影像的原阳县玉米和水稻分类时间窗选择[J].江苏农业学报,2023,(02):413.[doi:doi:10.3969/j.issn.1000-4440.2023.02.014]
　LI Chang-chun,ZHAI Wei-guang,WANG Chun-yang,et al.Time window selection of corn and rice classification in Yuanyang County based on Sentinel-1A image[J].,2023,(09):413.[doi:doi:10.3969/j.issn.1000-4440.2023.02.014]
[3]郭玉娣,李根,李春,等.基于多时相合成孔径雷达数据的水稻种植面积监测[J].江苏农业学报,2023,(05):1179.[doi:doi:10.3969/j.issn.1000-4440.2023.05.010]
　GUO Yu-di,LI Gen,LI Chun,et al.Rice planting area monitoring based on multi-temporal synthetic aperture radar (SAR) data[J].,2023,(09):1179.[doi:doi:10.3969/j.issn.1000-4440.2023.05.010]
[4]樊泳灼,李新国.湖滨绿洲棕漠土有机碳含量高光谱估算[J].江苏农业学报,2023,(06):1341.[doi:doi:10.3969/j.issn.1000-4440.2023.06.009]
　FAN Yong-zhuo,LI Xin-guo.Hyperspectral prediction of organic carbon content of brown desert soil in the lakeside oasis[J].,2023,(09):1341.[doi:doi:10.3969/j.issn.1000-4440.2023.06.009]
[5]钟怡琪,李家国,韩杰,等.基于哨兵影像与多特征优选的溧阳市上兴镇水稻识别[J].江苏农业学报,2023,(08):1688.[doi:doi:10.3969/j.issn.1000-4440.2023.08.008]
　ZHONG Yi-qi,LI Jia-guo,HAN Jie,et al.Identification of rice in Shangxing Town, Liyang City based on Sentinel image and multi-feature optimization[J].,2023,(09):1688.[doi:doi:10.3969/j.issn.1000-4440.2023.08.008]
[6]于天祥,樊红.基于Sentinel-2多时相遥感影像的冬小麦种植面积监测[J].江苏农业学报,2024,(09):1653.[doi:doi:10.3969/j.issn.1000-4440.2024.09.009]
　YU Tianxiang,FAN Hong.Remote sensing monitoring of winter wheat planting area based on multi-temporal Sentinel-2 imagery[J].,2024,(09):1653.[doi:doi:10.3969/j.issn.1000-4440.2024.09.009]

备注/Memo

备注/Memo:
收稿日期：2023-11-10基金项目：国家重点研发计划项目(2023YFB3905801)作者简介：罗凤宇(1998-),男,贵州黔南人,硕士研究生,主要研究方向为茶树遥感识别。(E-mail)20211210039@nuist.edu.cn通讯作者：谢勇,(E-mail) xieyong@nuist.edu.cn
更新日期/Last Update: 2024-11-17