基于Sentinel-2时序数据的广东省英德市茶园分类研究

doi:10.3964/j.issn.1000-0593(2024)04-1136-08

摘要
参考文献
相关文章 (15)

全文: PDF (8675 KB)
输出: BibTeX | EndNote (RIS)

摘要：茶叶是一种高附加值的经济作物，是我国南方山区乡村振兴的主要抓手。由于毁林种茶等破坏行为，导致森林资源破坏并引发水土流失等生态环境问题。快速准确获取茶园的空间分布对于政府监管和茶叶产业的规划发展至关重要。由于研究区天气多阴雨，茶园分布较为分散，与森林等植被光谱较为接近等原因，导致基于卫星影像提取茶园挑战性较大。为了摸清英德市的茶园空间分布，系统分析了中高分辨率的多光谱Sentinel-2影像数据，结合多时序多特征信息在茶园提取中的应用潜力。以英德市全境为研究区，选用2019年—2021年的9期Sentinel-2影像数据，详细分析了茶树生长的物候特征，进一步探究了茶园和其他地类在多时序中的特征变化，采用Relief算法对所有特征进行重要性排序。根据特征排序结果，选取特征权重值加权90%的特征因子，即7个植被指数特征和2个纹理特征，通过不同的组合排序构建了9种茶园分类场景，采用RF算法对所有分类场景进行精度评价，选取最佳分类场景，进一步探讨了RF分类算法和SVM分类算法对茶园提取的可行性。结果表明：(1）在进行英德市茶园提取时，2月和10月是采用多时相构造茶园多特征的最佳组合，可能因2月茶树处于萌芽期长出部分嫩绿的新叶易于和森林植被区分且在10月前后由于茶园进行了修剪其特征也较明显，因此两时相特征融合易于区分茶园。(2）RF分类方法与SVM分类方法相比，后者的精度较高，其总体精度达到91.56%，Kappa系数为0.89，生产者精度和用户精度分别为80.22%和84.56%。该研究为快速高效获取英德市茶园空间分布信息提供了一种高效的方法，同时为政府在进行茶叶产业规划、管理提供了数据支持。

关键词：茶园；Sentinel-2；时序特征；机器学习；分类

Abstract：Tea is a high value-added economic crop with extremely high economic value. It is the main starting point for rural revitalization in mountainous areas of China. However, due to destructive behaviors such as deforestation and planting tea, forest resources are destroyed, and ecological and environmental problems such as soil erosion are caused. Acquiring the spatial distribution of tea plantations quickly and accurately is very important for government supervision and the planning and development of the tea industry. However, due to the rainy weather in the study area and the scattered distribution of tea plantations, which are close to the spectrum of vegetation such as forests, the extraction based on satellite imagery has become a problem. Tea plantations are challenging. In order to find out the spatial distribution of tea plantations in Yingde, this paper systematically analyzes the application potential of medium and high-resolution multispectral Sentinel-2 image data combined with multi-time-series and multi-feature information in tea garden extraction. Taking the whole territory of Yingde as the research area, this paper selects 9 phases of Sentinel-2 image data from 2019 to 2021 to analyze the phenological characteristics of tea tree growth in detail and further explore the characteristics changes of tea plantations and other land types in multiple time series, using the Relief algorithm to sort the importance of all features. According to the result of feature sorting, the feature factors weighted by 90% of the feature weight value are selected, namely 7 vegetation index features and 2 texture features, and 9 kinds of tea garden classification scenes are constructed through different combination rankings, and the RF algorithm is used to evaluate the accuracy of all classification scenes. To select the best classification scene and further discuss the feasibility of the RF classification algorithm and SVM classification algorithm for tea garden extraction. The results show that: (1) When extracting tea plantations in Yingde, February and October are the best combinations to construct multiple characteristics of tea plantations using multi-temporal phases. (2) Compared with the SVM classification method, the RF classification method has high accuracy. Its overall accuracy reaches 91.56%, the Kappa coefficient is 0.89, and the producer accuracy and user accuracy are 80.22% and 84.56%, respectively. This study provides an efficient method for quickly and efficiently obtaining the spatial distribution information of tea plantations in Yingde and provides data support for the government to plan and manage the tea industry.

Key words：Tea plantation; Sentinel-2; Temporal features; Machine learning; Classification

收稿日期: 2022-09-02 修订日期: 2023-02-20

中图分类号:

S758.7

基金资助: 国家重点研发计划项目(2020YFD1100202)，北京市农林科学院2022年科研创新平台项目(PT2022-24), 北京市农林科学院博士后基金项目(2021-ZZ-002)资助

通讯作者: 任艳敏，刘玉 E-mail: renym@nercita.org.cn; liuyu@nercita.org.cn

作者简介: 陈盼盼，女，1990年生，北京市农林科学院信息技术研究中心博士研究生 e-mail: jeno.chen@foxmail.com

引用本文:

陈盼盼，任艳敏，赵春江，李存军，刘玉. 基于Sentinel-2时序数据的广东省英德市茶园分类研究[J]. 光谱学与光谱分析, 2024, 44(04): 1136-1143.
CHEN Pan-pan, REN Yan-min, ZHAO Chun-jiang, LI Cun-jun, LIU Yu. Research on the Classification of Yingde Tea Plantations Based on Time Series Sentinel-2 Images. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(04): 1136-1143.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2024)04-1136-08 或 https://www.gpxygpfx.com/CN/Y2024/V44/I04/1136

[1] Su S, Chen W, Jing L, et al. Land Use Policy, 2017, 66: 183.
[2] ZHOU Shu-dong, WU Xin-lin（周曙东, 吴新林）. Journal of Tea Communication（茶叶通讯），2020, 47(3): 6.
[3] Liu S, Yin Y, Liu X, et al. Ecological Modelling, 2017, 353: 129.
[4] Su S，Zhou X，Wan C，et al. Land Use Policy，2016，50: 379.
[5] Zhu J, Pan Z, Wang H, et al. Sensors, 2019, 19(9): 2087.
[6] Chuang Y C M, Shiu Y S. Sensors, 2016, 16(5): 594.
[7] Dihkan M, Guneroglu N, Karsli F, et al. International Journal of Remote Sensing, 2013, 34(23): 8549.
[8] XU Wei-yan, SUN Rui, JIN Zhi-feng（徐伟燕, 孙睿, 金志凤）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2016, 32(S1): 161.
[9] MA Chao, YANG Fei, WANG Xue-cheng（马超, 杨飞, 王学成）. Remote Sensing for Land & Resources（国土资源遥感）, 2019, 31(1): 141.
[10] Das A C. Remote Sensing, 2020, 12(24): 25.
[11] YANG Yan-kui, CHEN Yun-zhi, WU Bo, et al（杨艳魁, 陈芸芝, 吴波, 等）. Jiangsu Agricultural Sciences（江苏农业科学）, 2019, 47(2): 210.
[12] TIAN Ying, CHEN Zhuo-qi, HUI Feng-ming et al（田颖, 陈卓奇, 惠凤鸣, 等）. Journal of Beijing Normal University(Natural Science)[北京师范大学学报(自然科学版)], 2019, 55(1): 57.
[13] Kim H O, Yeom J M. International Journal of Remote Sensing, 2014, 35(19-20): 7046.
[14] Rim M, Makram A. International Journal of Remote Sensing, 2020, 41(23): 8986.
[15] Markus I, Francesco V, Clement A. Remote Sensing, 2016, 8(3): 166.
[16] Jesús Soriano-Gonzalez, Eduard Angelats, Maite Martínez-Eixarch, et al. Field Crops Research, 2022, 281: 108507.
[17] Yu Shen, Xiaoyang Zhang, Zhengwei Yang. ISPRS Journal of Photogrammetry and Remote Sensing，2022, 186：55.
[18] Jie Wang, Xiangming Xiao, Luo Liu, et al. Remote Sensing of Environment, 2020, 247：111951.
[19] LI Long-wei, LI Nan, LU Deng-sheng（李龙伟, 李楠, 陆灯盛）. Journal of Zhejiang A&F University（浙江农林大学学报）, 2019, 36(5): 841.
[20] ZHAO Xiao-qing, WANG Ping, JING Lin-hai, et al（赵晓晴, 王萍, 荆林海, 等）. Science of Surveying and Mapping（测绘科学）, 2020, 45(6): 80.
[21] LI Dan-xia, FANG Wei, LIANG Jun-fen, et al（李丹霞, 方伟, 梁俊芬, 等）. Tropical Agricultural Engineering（热带农业工程）, 2019, 43(6): 73.
[22] CHEN Hui-ying, CAO Jun-xi, SUN Shi-li, et al（陈慧英, 操君喜, 孙世利, 等）. Guangdong Agricultural Sciences（广东农业科学）, 2020, 47(11): 209.
[23] Xia Z, Wang X, Cao G, et al. Journal of Agricultural Science, 2017, 9(4): 116.
[24] Fern R R, Foxley E A, Bruno A, et al. Ecological Indicators, 2018, 94: 16.
[25] WU Jing, LÜ Yu-na, LI Chun-bin, et al（吴静, 吕玉娜, 李纯斌, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报）, 2019, 50(9): 194.
[26] ZHOU Jing-ping，LI Cun-jun，SHI Lei-gang，et al（周静平, 李存军, 史磊刚, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报），2016, 47(9): 318.
[27] LIN Tao, ZHANG Da, WANG Jian-jun（林涛, 张达, 王建君）. Computer Simulation（计算机仿真）, 2021, 38(9): 414.
[28] WANG Geng-ze, JIN Hai-liang, GU Xiao-he, et al（王庚泽, 靳海亮, 顾晓鹤, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报），2021, 52(2): 199.
[29] Zhang F, Yang M J. Remote Sensing of Environment, 2020, 251: 112105.
[30] Musbah H, Aly H H, Little T A. Electric Power Systems Research, 2021, 199: 107436.
[31] WANG Bin, HE Bing-hui, LIN Na, et al（王斌, 何丙辉, 林娜, 等）. Journal of Jilin University (Engineering and Technology Edition)[吉林大学学报(工学版)], 2022, 52(7): 1719.
[32] BAI Jia, SUN Rui, ZHANG He-lin, et al(柏佳, 孙睿, 张赫林, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2021, 37(14): 179.
[33] Chen Panpan, Li Cunjun, Chen Shilin, et al. Remote Sensing，2022, 14, 2412.