地球信息科学学报 ›› 2020, Vol. 22 ›› Issue (12): 2445-2455.doi: 10.12082/dqxxkx.2020.200338
• 遥感科学与应用技术 • 上一篇
杨丹1(), 周亚男1,*(
), 杨先增1, 郜丽静2, 冯莉1
收稿日期:
2020-06-30
修回日期:
2020-09-15
出版日期:
2020-12-25
发布日期:
2021-02-25
通讯作者:
周亚男
E-mail:danyanggis@163.com;zhouyn@hhu.edu.cn
作者简介:
杨 丹(1996— ),女,安徽枞阳人,硕士生,主要从事遥感植被分类研究。E-mail: 基金资助:
YANG Dan1(), ZHOU Yanan1,*(
), YANG Xianzeng1, GAO Lijing2, FENG Li1
Received:
2020-06-30
Revised:
2020-09-15
Online:
2020-12-25
Published:
2021-02-25
Contact:
ZHOU Yanan
E-mail:danyanggis@163.com;zhouyn@hhu.edu.cn
Supported by:
摘要:
植被分类是森林资源调查与动态监测的基础与前提。当前植被分类研究大都利用光学遥感影像,然而,光学遥感成像易受到云雨覆盖的影响,难以构建完整时间序列,植被分类精度有限。微波遥感具有全天时全天候、时间序列完整的优势,在植被调查与分析中具有巨大的应用潜力。本文利用2018年Sentinel-1A微波遥感时间序列数据和深度循环网络方法,对秦岭太白山区的森林植被进行分类制图。首先利用Sentinel-2光学影像与数字高程数据对研究区进行多尺度分割;然后将处理后的时间序列Sentinel-1A数据空间叠加到分割地块上,构建地块的多元时间序列曲线;最后利用深度循环网络提取与学习多元时间序列的时序特征并分类。实验结果表明:① 与传统机器学习方法(如RF、SVM)相比,本文提出的深度循环网络方法的分类精度提高10%以上;② 在Sentinel-1A微波极化特征组合中VV+VH表现最好,与VV+VH+VV/VH极化特征组合的精度相近;③ 使用全年的时间影像构建时间序列分类精度最高,达到82%。研究表明,利用深度循环网络与时间序列Sentinel-1A数据的方法能够有效提高植被分类的精度,从数据源与分类方法上为森林植被分类研究提供了新的思路。
杨丹, 周亚男, 杨先增, 郜丽静, 冯莉. LSTM支持下时序Sentinel-1A数据的太白山区植被制图[J]. 地球信息科学学报, 2020, 22(12): 2445-2455.DOI:10.12082/dqxxkx.2020.200338
YANG Dan, ZHOU Yanan, YANG Xianzeng, GAO Lijing, FENG Li. Vegetation Mapping in Taibai Mountain Area Supported by LSTM with Time Series Sentinel-1A Data[J]. Journal of Geo-information Science, 2020, 22(12): 2445-2455.DOI:10.12082/dqxxkx.2020.200338
表1
Sentinel-1A SAR数据列表"
序号 | 影像获取时间 | 极化方式 | 空间分辨率 | 序号 | 影像获取时间 | 极化方式 | 空间分辨率 |
---|---|---|---|---|---|---|---|
D1 | 2018-01-11 | VV+VH | 5 m×20 m | D16 | 2018-07-10 | VV+VH | 5 m×20 m |
D2 | 2018-01-23 | D17 | 2018-07-22 | ||||
D3 | 2018-02-04 | D18 | 2018-08-03 | ||||
D4 | 2018-02-16 | D19 | 2018-0815 | ||||
D5 | 2018-02-28 | D20 | 2018-08-27 | ||||
D6 | 2018-03-12 | D21 | 2018-09-08 | ||||
D7 | 2018-03-24 | D22 | 2018-09-20 | ||||
D8 | 2018-04-05 | D23 | 2018-10-02 | ||||
D9 | 2018-04-17 | D24 | 2018-10-14 | ||||
D10 | 2018-04-29 | D25 | 2018-10-26 | ||||
D11 | 2018-05-11 | D26 | 2018-11-07 | ||||
D12 | 2018-05-23 | D27 | 2018-11-19 | ||||
D13 | 2018-06-04 | D28 | 2018-12-01 | ||||
D14 | 2018-06-16 | D29 | 2018-12-13 | ||||
D15 | 2018-06-28 | D30 | 2018-12-25 |
表3
RF、SVM与LSTM3种分类方法的精度评价"
类别 | SVM | RF | LSTM | ||||||
---|---|---|---|---|---|---|---|---|---|
UA | PA | F1 | UA | PA | F1 | UA | PA | F1 | |
阔叶林 | 68.75 | 72.24 | 70.45 | 69.43 | 75.24 | 72.22 | 83.53 | 84.67 | 84.10 |
针叶林 | 66.32 | 67.43 | 66.87 | 71.18 | 74.14 | 72.63 | 84.31 | 82.59 | 83.44 |
针阔混交林 | 64.03 | 63.45 | 63.74 | 65.29 | 73.34 | 69.08 | 80.04 | 81.27 | 80.65 |
灌丛 | 65.14 | 62.78 | 63.94 | 68.25 | 68.74 | 68.49 | 81.14 | 78.96 | 80.04 |
草甸 | 64.91 | 62.36 | 63.61 | 67.74 | 69.37 | 68.55 | 79.46 | 79.35 | 79.40 |
总体精度 | 65.57 | 71.39 | 82.45 | ||||||
Kappa系数 | 61.34 | 69.58 | 80.01 |
[1] | 朱晓勤, 刘康, 李建国, 等. GIS支持下的秦岭山地植被分布与环境梯度关系研究[J]. 水土保持研究, 2009,16(2):169-175. |
[ Zhu X Q, Liu K, Li J G, et al. Analysis on vegetation-environment gradient correlation in Qinling Mountain based on GIS[J]. Research of Soil and Water Conservation, 2009,16(2):169-175. ] | |
[2] | 竺可桢. 中国的亚热带[J]. 科学通报, 1958(17):524-528. |
[ Zhu K Z. Chinese subtropics[J]. Chinese Science Bulletin., 1958(17):524-528. ] | |
[3] |
张百平. 中国南北过渡带研究的十大科学问题[J]. 地理科学进展, 2019,38(3):305-311.
doi: 10.18306/dlkxjz.2019.03.001 |
[ Zhang B P. Ten major scientific issues concerning the study of China's north-south transitional zone[J]. Progress in Geography, 2019,38(3):305-311. ] | |
[4] |
崔一娇, 朱琳, 赵力娟. 基于面向对象及光谱特征的植被信息提取与分析[J]. 生态学报, 2013,33(3):867-875.
doi: 10.5846/stxb201204110510 |
[ Cui Y J, Zhu L, Zhao L J. Abstraction and analysis of vegetation information based on object-oriented and spectra features[J]. Acta Ecologica Sinica, 2013,33(3):867-875. ] | |
[5] |
贾明明, 任春颖, 刘殿伟, 等. 基于环境星与MODIS时序数据的面向对象森林植被分类[J]. 生态学报, 2014,34(24):7167-7174.
doi: 10.5846/stxb201310112438 |
[ Jia M M, Ren C Y, Liu D W. Object-oriented forest classification based on combination of HJ-1 CCD and MODIS-NDVI data[J]. Acta Ecologica Sinica, 2014,34(24):7167-7174. ] | |
[6] | 杨存建, 周其林, 任小兰, 等. 基于多时相MODIS数据的四川省森林植被类型信息提取[J]. 自然资源学报, 2014,29(3):507-515. |
[ Yang C J, Zhou Q L, Ren X L, et al. Extracting forest vegetation types from multi-temporal MODIS imagery in Sichuan Province[J]. Journal of Natural Resources, 2014,29(3):507-515. ] | |
[7] |
胡秀娟, 徐涵秋, 黄绍霖, 等. WorldView-2近红外光谱波段反演马尾松植被信息的比较研究[J]. 地球信息科学学报, 2016,18(4):537-543.
doi: 10.3724/SP.J.1047.2016.00537 |
[ Hu X J, Xu H Q, Huang S L, et al. Comparison between the two near infrared bands of WorldView-2 imagery in their applications in Pinus Massoniana Forest[J]. Journal of Geo-information Science, 2016,18(4):537-543. ] | |
[8] | 李丹, 柯樱海, 宫辉力, 等. 基于高分辨率遥感影像的城市典型乔木树种分类研究[J]. 地理与地理信息科学, 2016,32(1):84-89,127. |
[ Li D, Ke Y H, Gong H L, et al. Urban tree species classification with machine learning classifier using WorldView-2 imagery[J]. Geography and Geo-Information Science, 2016,32(1):84-89,127. ] | |
[9] | 张俊瑶, 姚永慧, 索南东主, 等. 基于垂直带谱的太白山区山地植被遥感信息提取[J]. 地理信息科学学报, 2019,21(8):1284-1294. |
[ Zhang J Y, Yao Y H, Suonan D Z, et al. Mapping of mountain vegetation in Taibai Mountain based on mountain altitudinal belts with remote sensing[J]. Journal of Geo-information Science, 2019,21(8):1284-1294. ] | |
[10] |
别强, 何磊, 赵传燕. 基于影像融合和面向对象技术的植被信息提取研究[J]. 遥感技术与应用, 2014,29(1):164-171.
doi: 10.11873/j.issn.1004-0323.2014.1.0164 |
[ Bie Q, He L, Zhao C Y. Study on vegetation information extraction based on object-oriented image analysis[J]. Remote Sensing Technology and Application, 2014,29(1):164-171. ] | |
[11] | Dostalova A, Wagner W, Milenkovic M, et al. Annual seasonality in Sentinel-1 signal for forest mapping and forest type classification[J]. International Journal of Remote Sensing, 2018,39(21):7738-7760. |
[12] | Grabska E, Hostert P, Pflugmacher D, et al. Forest stand species mapping using the Sentinel-2 time series[J]. Remote Sens-Basel, 2019,11(10):1197. |
[13] | 冯莉, 李柳华, 郭松, 等. HJ-1A NDVI与MODIS NDVI时间序列提取植被物候特征对比研究[J]. 遥感技术与应用, 2016,31(6):1158-1166. |
[ Feng L, Li L H, Guo S, et al. Comparative study on extracting vegetation phenology information from HJ-1A NDVI and MODIS NDVI time series[J]. Remote Sensing Technology and Application, 2016,31(6):1158-1166. ] | |
[14] | 郭文婷, 张晓丽. 基于Sentinel-2时序多特征的植被分类[J]. 浙江农林大学学报, 2019,36(5):849-856. |
[ Guo W T, Zhang X L. Vegetation classification based on a multi-feature Sentinel-2 time series[J]. Journal of Zhejiang A & F University, 2019,36(5):849-856. ] | |
[15] | 郝莹莹, 罗小波, 仲波, 等. 基于植被分区的中国植被类型分类方法[J]. 遥感技术与应用, 2017,32(2):315-323. |
[ Hao Y Y, Luo X B, Zhong B, et al. Methods of the national vegetation classification based on vgetation partition[J]. Remote Sensing Technology and Application, 2017,32(2):315-323. ] | |
[16] | Li H, Zhang C, Zhang S, et al. A hybrid OSVM-OCNN method for crop classification from fine spatial resolution remotely sensed imagery[J]. Remote Sens-Basel, 2019,11(20):2370. |
[17] | Tomppo E, Antropov O, Praks J. Cropland classification using Sentinel-1 time series: Methodological performance and prediction uncertainty assessment[J]. Remote Sens-Basel, 2019,11(21):2480. |
[18] | Zhou Yn, Luo J, Feng L, et al. Long-short-term-memory-based crop classification using high-resolution optical images and multi-temporal SAR data[J]. GIScience & Remote Sensing, 2019,56(8):1170-1191. |
[19] | Mahdavi S, Salehi B, Amani M, et al. A dynamic classification scheme for mapping spectrally similar classes: Application to wetland classification[J]. International Journal of Applied Earth Observation and Geoinformation, 2019,83:101914. |
[20] | Mohammadimanesh F, Salehi B, Mahdianpari M, et al. Full and simulated compact polarimetry sar responses to canadian wetlands: Separability analysis and classification[J]. Remote Sens-Basel, 2019,11(5):516. |
[21] | Mohammadimanesh F, Salehi B, Mahdianpari M, et al. Multi-temporal, multi-frequency, and multi-polarization coherence and SAR backscatter analysis of wetlands[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018,142:78-93. |
[22] | Grimaldi S, Xu J, Li Y, et al. Flood mapping under vegetation using single SAR acquisitions[J]. Remote Sensing of Environment, 2020,237:111582. |
[23] | Parrens M, Al Bitar A, Frappart F, et al. High resolution mapping of inundation area in the Amazon basin from a combination of L-band passive microwave, optical and radar datasets[J]. International Journal of Applied Earth Observation and Geoinformation. 2019,81:58-71. |
[24] | Tsyganskaya V, Martinis S, Marzahn P. Flood monitoring in vegetated areas using multitemporal Sentinel-1 data: Impact of time series features[J]. Water, 2019,11(9):1938. |
[25] | Erinjery J J, Singh M, Kent R. Mapping and assessment of vegetation types in the tropical rainforests of the Western Ghats using multispectral Sentinel-2 and SAR Sentinel-1 satellite imagery[J]. Remote Sensing of Environment, 2018,216:345-354. |
[26] | Yu Y, Li M, Fu Y. Forest type identification by random forest classification combined with SPOT and multitemporal SAR data[J]. Journal of forestry research, 2018,29(5):1407-1414. |
[27] | Zhang H, Wang T, Liu M, et al. Potential of combining optical and dual polarimetric sar data for improving mangrove species discrimination using rotation forest[J]. Remote Sens-Basel, 2018,10(3):467. |
[28] | Sakarya, Ufuk, Demirpolat, et al. SAR image time series analysis in vegetation classification using global and local information based linear discriminant analysis[C]. 25th Signal Processing and Communications Applications Conference (SIU), Antalya, Turkey, 2017. |
[29] | Ihrig, Ramona, Kuny, et al. Analysis of phenological changes of high vegetation in amplitude images of SAR time series[C]. 17th SPIE Conference on Earth Resources and Environmental Remote Sensing/GIS Applications, Warsaw, Poland, 2017. |
[30] | Julie B, Sebastien R, Samuel C, et al. TerraSAR-X dual-pol time-series for mapping of wetland vegetation[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015,107:90-98. |
[31] | 曹林林, 李海涛, 韩颜顺, 等. 卷积神经网络在高分遥感影像分类中的应用[J]. 测绘科学, 2016,41(9):170-175. |
[ Cao L L, Li H T, Han Y S, et al. Application of convolutional neural networks in classification of high resolution remote sensing imagery[J]. Science of Surveying and Mapping, 2016,41(9):170-175. ] | |
[32] | 刘浩, 骆剑承, 黄波, 等. 基于特征压缩激活Unet网络的建筑物提取[J]. 地球信息科学学报, 2019,21(11):1779-1789. |
[ Liu H, Luo J C, Huang B, et al. Building extraction based on SE-Unet[J]. Journal of Geo-information Science, 2019,21(11):1779-1789. ] | |
[33] | 吴田军, 骆剑承, 夏列钢, 等. 迁移学习支持下的遥感影像对象级分类样本自动选择方法[J]. 测绘学报, 2014,43(9):908-916. |
[ Wu T J, Luo J C, Xia L G, et al. An automatic sample collection method for object-oriented classification of remotely sensed imageries based on transfer learning[J]. Acta Geodaetica et Cartographica Sinica, 2014,43(9):908-916. ] | |
[34] | Hofmann P. Detecting buildings and roads from IKONOS data using additional elevation information[J]. GeoBIT/GIS, 2001,6(1):28-33. |
[35] |
Dragut L, Tiede D, Levick S R. ESP: A tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data[J]. International Journal of Geographical Information Science, 2010,24(6):859-871.
doi: 10.1080/13658810903174803 |
[36] | Agrawal S, Joshi P K, Shukla Y, et al. Spot Vegetation multi temporal data for classifying vegetation in south central Asia[J]. Current Science, 2003,84(11):1440-1448. |
[37] | Aurdal L, Huseby R B, Eikvil L, et al. Use of hidden Markov models and phenology for multitemporal satellite image classification: applications to mountain vegetation classification[A]. 2005 International Workshop on the Analysis of Multi-Temporal Remote Sensing Images[C]. 2005. 220-224. |
[38] | Wu F, Wang C, Zhang H, et al. Rice crop monitoring in South China With RADARSAT-2 quad- polarization SAR data[J]. IEEE Geoscience & Remote Sensing Letters, 2011,8(2):196-200. |
[1] | 谢聪慧, 吴世新, 张晨, 孙文涛, 何海芳, 裴韬, 罗格平. 基于谱系聚类的全球各国新冠疫情时间序列特征分析[J]. 地球信息科学学报, 2021, 23(2): 236-245. |
[2] | 刘明杰, 徐卓揆, 郜允兵, 杨晶, 潘瑜春, 高秉博, 周艳兵, 周万鹏, 王凌. 基于机器学习的稀疏样本下的土壤有机质估算方法[J]. 地球信息科学学报, 2020, 22(9): 1799-1813. |
[3] | 陈昂, 杨秀春, 徐斌, 金云翔, 张文博, 郭剑, 邢晓语, 杨东. 基于面向对象与深度学习的榆树疏林识别方法研究[J]. 地球信息科学学报, 2020, 22(9): 1897-1909. |
[4] | 刘新, 赵宁, 郭金运, 郭斌. 基于LSTM神经网络的青藏高原月降水量预测[J]. 地球信息科学学报, 2020, 22(8): 1617-1629. |
[5] | 李婉, 牛陆, 陈虹, 吴骅. 基于随机森林算法的地表温度鲁棒降尺度方法[J]. 地球信息科学学报, 2020, 22(8): 1666-1678. |
[6] | 毛亚萍, 房世峰. 基于机器学习的参考作物蒸散量估算研究[J]. 地球信息科学学报, 2020, 22(8): 1692-1701. |
[7] | 吴志峰, 骆剑承, 孙营伟, 吴田军, 曹峥, 刘巍, 杨颖频, 王玲玉. 时空协同的精准农业遥感研究[J]. 地球信息科学学报, 2020, 22(4): 731-742. |
[8] | 郭云开, 张晓炯, 许敏, 刘雨玲, 钱佳, 章琼. 路域植被等效水厚度估算模型研究[J]. 地球信息科学学报, 2020, 22(2): 308-315. |
[9] | 郑诗晨, 盛业华, 吕海洋. 基于粒子滤波的行车轨迹路网匹配方法[J]. 地球信息科学学报, 2020, 22(11): 2109-2117. |
[10] | 周玲, 李云梅, 赵焕, 房晟忠, 颜翔, 吕恒. 滇池湖滨带植被时空变化及其对滇池总磷浓度的影响[J]. 地球信息科学学报, 2020, 22(10): 1946-1958. |
[11] | 汪雪淼, 孟庆岩, 赵少华, 李娟, 张琳琳, 陈旭. GF-2在城市绿地分类及景观格局度量中的应用[J]. 地球信息科学学报, 2020, 22(10): 1971-1982. |
[12] | 高亮, 杜鑫, 李强子, 王红岩, 张源, 王思远. 融合土地覆盖和土壤水分产品的近地表空气温度空间化方法[J]. 地球信息科学学报, 2020, 22(10): 2023-2037. |
[13] | 叶杰, 孟凡晓, 白潍铭, 张斌, 郑金明. “四同”条件下周口城区高分一号遥感影像分类对比研究[J]. 地球信息科学学报, 2020, 22(10): 2088-2097. |
[14] | 方志祥, 倪雅倩, 黄守倩. 顾及上网行为特征的手机用户停留行为预测方法[J]. 地球信息科学学报, 2020, 22(1): 136-144. |
[15] | 宋关福, 卢浩, 王晨亮, 胡辰璞, 黄科佳. 人工智能GIS软件技术体系初探[J]. 地球信息科学学报, 2020, 22(1): 76-87. |
|