Research on Interpretation of Ramie Lodging Information Based on Unmanned Aerial Vehicles

doi:10.13304/j.nykjdb.2022.1066

Abstract

Abstract:

The most common damage to ramie tramet cultivation is stem loading. Traditional monitoring methods have drawbacks such as being time-consuming and inefficient. A method for obtaining ramie lodging information was investigated by unmanned aerial vehicles （UAV） in this study. Firstly， the canopy orthophoto and digital surface model （DSM） of ramie were created using Pix4D Mapper software. Then， the spectral， textural， and shape features of the canopy were extracted from the DSM， along with the canopy height index. Finally， a combination of 3 machine learning algorithms was used to createa classification model for normal and lodging canopies. The results showed that the DSM-based extracted plant height information could effectively replace the actual measured plant height in the field， with a model R² of 0.899. The spectral， textural， shape， and height characteristics of fallen and normal ramets differed. The support vector machine and decision tree models outperformed the other learning algorithms， achieving 99% accuracy and efficiently identifying normal/lodging ramie plots. Above results provided technical assistance for accurate and rapid assessment of crop lodging.

Key words: ramie, lodging, unmanned aerial vehicles（UAV）, visible light camera, digital surface model（DSM）

摘要：

茎杆倒伏是苎麻三麻培育中最常见的灾害，传统的监测方法具有耗时耗力、不及时等局限性。提出了一种基于无人机航拍获取苎麻倒伏信息的方法，首先利用Pix4D Mapper软件生成苎麻的冠层正射影像和数字表面模型（digital surface model，DSM），基于正射影像提取苎麻光谱、纹理及形状特征，基于DSM提取苎麻株高指标，最后结合3种机器学习算法构建正常/倒伏苎麻分类模型。结果表明，基于DSM提取的株高信息可以有效代替大田实测株高，模型R²为0.899。倒伏和正常苎麻在光谱、纹理、形状及株高特征上具有差异。在3种机器学习算法中，支持向量机和决策树模型的性能最好，准确率达到99%，能够高效地识别苎麻倒伏地块。以上研究结果为准确、快速评估作物倒伏情况提供了技术支撑。

关键词: 苎麻, 倒伏, 无人机, 可见光相机, 数字表面模型

CLC Number:

S127

Wei WANG, Hongyu FU, Jianning LU, Yunkai YUE, Ruifang YANG, Guoxian CUI, Wei SHE. Research on Interpretation of Ramie Lodging Information Based on Unmanned Aerial Vehicles[J]. Journal of Agricultural Science and Technology, 2024, 26(3): 91-97.

王薇, 付虹雨, 卢建宁, 岳云开, 杨瑞芳, 崔国贤, 佘玮. 基于无人机航拍的苎麻倒伏信息解译研究[J]. 中国农业科技导报, 2024, 26(3): 91-97.

Figures/Tables 6

Fig. 1 Location of test plot

Table 1 Characterization and calculation

特征分类 Classification	特征指标 Indicator	计算 Calculation
光谱 Optical spectrum	红色标准值 Red	r=R/（R+G+B）
	绿色标准值 Green	g=G/（R+G+B）
	蓝色标准值 Blue	b=B/（R+G+B）
	绿叶植被指数 Green Leaf Index（GLA）	GLA=（2G-R-B）/（2G+R+B）
	超绿植被指数 Excess Green Index（ExG）	ExG=2G-R-B
纹理 Texture	灰度共生矩阵同质度 GLCM homogeneity	H= $∑ i ∑ j 1 1 + i + j 2 p (i, j)$
	灰度共生矩阵对比度 GLCM contrast	Con= $∑ i ∑ j (i - j) 2 p (i, j)$
	灰度共生矩阵熵 GLCM entropy	Ent= $∑ i ∑ j p (i, j) l o g p (i, j)$
	灰度共生矩阵均值 GLCM mean	M= $∑ i, j = 0 N - 1 P i, j / N 2$
形状 Shape	边界指数 Border index（Bd）	Bd=e/L_minBdRd
	圆度 Roundness（Rd）	Rd=Rd_max-Rd_min
	紧凑度 Compactness	C=S_minBdRd/A
	密度 Density	D=S/R₁

Table 1 Characterization and calculation

特征分类 Classification	特征指标 Indicator	计算 Calculation
光谱 Optical spectrum	红色标准值 Red	r=R/（R+G+B）
	绿色标准值 Green	g=G/（R+G+B）
	蓝色标准值 Blue	b=B/（R+G+B）
	绿叶植被指数 Green Leaf Index（GLA）	GLA=（2G-R-B）/（2G+R+B）
	超绿植被指数 Excess Green Index（ExG）	ExG=2G-R-B
纹理 Texture	灰度共生矩阵同质度 GLCM homogeneity	H= $∑ i ∑ j 1 1 + i + j 2 p (i, j)$
	灰度共生矩阵对比度 GLCM contrast	Con= $∑ i ∑ j (i - j) 2 p (i, j)$
	灰度共生矩阵熵 GLCM entropy	Ent= $∑ i ∑ j p (i, j) l o g p (i, j)$
	灰度共生矩阵均值 GLCM mean	M= $∑ i, j = 0 N - 1 P i, j / N 2$
形状 Shape	边界指数 Border index（Bd）	Bd=e/L_minBdRd
	圆度 Roundness（Rd）	Rd=Rd_max-Rd_min
	紧凑度 Compactness	C=S_minBdRd/A
	密度 Density	D=S/R₁

Table 2 HDSM and actual ramie height

指标

Index

均值

Mean/m

标准差

Standard deviation

标准误差均值

Standard error mean

株高数字表面模型

H_DSM

1.691

0.465

0.039 7

实际株高

Actual height

1.675

0.174

0.015 0

Fig. 2 Differences in lodging and normal ramie characteristics

Table 3 Model results of three machine learning

模型 Model	精确率 Precision		召回率 Recall		F1分数 F1 score		准确率 Accuracy/%
模型 Model	训练集 Train	测试集 Validation	训练集 Train	测试集 Validation	训练集 Train	测试集 Validation	准确率 Accuracy/%
随机森林 Random forest	1.00	0.98	0.67	1.00	0.80	0.99	98
支持向量机 Support vector machine	1.00	0.99	0.83	1.00	0.91	1.00	99
决策树 Decision tree	0.86	1.00	1.00	0.99	0.92	1.00	99

Table 4 Biomass of normal ramie and lodging ramie

类型 Type	地上部生物产量 Aboveground biological yield（kg·m^-2）
类型 Type	最大值 Max	最小值 Min	均值 Mean	标准差 Standard deviation
正常苎麻 Normal ramie	5.34	0.41	2.372	0.909
倒伏苎麻 Lodging ramie	5.13	1.32	2.184	1.388
严重倒伏苎麻 Severe lodging ramie	1.46	0.79	1.199	0.258

References 19

1	刘利华,金再欣,刘小丽,等.水稻倒伏对产量影响的试验结果分析[J].中国稻米,2009(2):19-21.
2	张建,谢田晋,杨万能,等.近地遥感技术在大田作物株高测量中的研究现状与展望[J].智慧农业,2021,3(1):1-15.
	ZHANG J, XIE T J, YANG W N,et al.. Research status and prospect on height estimation of field crop using near-field remote sensing technology [J]. Smart Agric., 2021,3(1):1-15.
3	SU W, ZHANG M, BIAN D,et al.. Phenotyping of corn plants using unmanned aerial vehicle (UAV) images [J/OL]. Remote Sens.,2019,11(17):2021 [2022-11-06]. .
4	XUE J, GOU L, ZAO Y S,et al.. Effects of light intensity within the canopy on maize lodging [J]. Field Crops Res.,2016,188:133-141.
5	周龙飞.基于多源遥感数据的玉米倒伏灾情监测研究[D].青岛:山东科技大学,2020.
	ZHOU L F. Monitoring of maize lodging disaster based on multi-source remote sensing data [D]. Qingdao: Shandong University of Science and Technology,2020.
6	王猛,隋学艳,梁守真,等.夏玉米倒伏模拟试验及遥感监测[J].测绘科学,2017,42(8):137-141.
	WANG M, SUI X Y, LIANG S Z,et al.. Simulation test and remote sensing monitoring of summer corn lodging [J]. Sci. Survey. Map., 2017,42(8):137-141.
7	田明璐,班松涛,袁涛,等.基于低空无人机多光谱遥感的水稻倒伏监测研究[J].上海农业学报,2018,34(6):88-93.
	TIAN M L, BAN S T, YUAN T,et al.. Monitoring of lodged rice using low-altitude UAV based multispectral image [J]. Acta Agric. Shanghai, 2018,34(6):88-93.
8	陆洲,徐飞飞,罗明,等.倒伏水稻特征分析及其多光谱遥感提取方法研究[J].中国生态农业学报,2021,29(4):751-761.
	LU Z, XU F F, LUO M,et al.. Characteristic analysis of lodging rice and study of the multi-spectral remote sensing extraction method [J]. Chin. J. Eco-Agric., 2021,29(4):751-761.
9	樊鸿叶,李姚姚,卢宪菊,等.基于无人机多光谱遥感的春玉米叶面积指数和地上部生物量估算模型比较研究[J].中国农业科技导报,2021,23(9):112-120.
	FAN H Y, LI Y Y, LU X J,et al.. Comparative analysis of LAI and above-ground biomass estimation models based on UAV multispectral remote sensing [J]. J. Agric. Sci. Technol.,2021,23(9):112-120.
10	胡琪,武红旗,轩俊伟,等.基于无人机DSM的小麦倒伏信息提取[J].安徽农业科学,2020,48(16):227-231.
	HU Q, WU H Q, XUAN J W,et al.. Information extraction of wheat lodging based on DSM by UAV remote sensing [J]. J. Anhui Agric. Sci.,2020,48(16):227-231.
11	董锦绘,杨小冬,高林,等.基于无人机遥感影像的冬小麦倒伏面积信息提取[J].黑龙江农业科学,2016(10):147-152.
	DONG J H, YANG X D, GAO L,et al.. Information extraction of winter wheat lodging area based on UAV remote sensing image [J]. Heilongjiang Agric. Sci., 2016(10):147-152.
12	赵立成,段玉林,史云,等.基于无人机DSM的小麦倒伏识别方法[J].中国农业信息,2019,31(4):36-42.
	ZHAO L C, DUAN Y L, SHI Y,et al.. Wheat lodging identification using DSM by drone [J]. Chin. Agric. Inform.,2019,31(4):36-42.
13	PÉREZHARGUINDEGUY N, DÍAZ S, GARNIER E,et al.. New handbook for standardised measurement of plant functional traits worldwide [J/OL]. Australian J. Botany,2013,61: 12225[2022-11-06]. .
14	宋茜.农作物空间分布信息提取及其时空格局变化分析研究[D].北京:中国农业科学院,2018.
	SONG Q. Crop mapping and spatio-temporal analysis of agricultural land-use [D]. Beijing: Chinese Academy of Agricultural Sciences,2018.
15	梁永检,吴文志,施泽升,等.基于无人机RGB遥感的甘蔗株高估测[J].作物杂志,2023(1):226-232.
	LIANG Y J, WU W Z, SHI Z S,et al.. Estimation of sugarcane plant height based on UAV RGB remote sensing [J]. Crops,2023(1):226-232.
16	赵静,闫春雨,杨东建,等.基于无人机多光谱遥感的台风灾后玉米倒伏信息提取[J].农业工程学报,2021,37(24):56-64.
	ZHAO J, YAN C Y, YANG D J,et al.. Extraction of maize lodging information after typhoon based on UAV multispectral remote sensing [J]. Trans. Chin. Soc. Agric. Eng.,2021,37(24):56-64.
17	李宗南,陈仲新,王利民,等.基于小型无人机遥感的玉米倒伏面积提取[J].农业工程学报,2014,30(19):207-213.
	LI Z N, CHEN Z X, WANG L M,et al.. Area extraction of maize lodging based on remote sensing by small unmanned aerial vehicle [J]. Trans. Chin. Soc. Agric. Eng.,2014,30(19):207-213.
18	毛智慧,邓磊,赵晓明,等.利用无人机遥感提取育种小区玉米倒伏信息[J].中国农学通报,2019,35(3):62-68.
	MAO Z H, DENG L, ZHAO X M,et al.. Extraction of maize lodging in breeding plot based on UAV remote sensing [J]. Chin. Agric. Sci. Bull.,2019,35(3):62-68.
19	申华磊,苏歆琪,赵巧丽,等.基于深度学习的无人机遥感小麦倒伏面积提取方法[J].农业机械学报,2022,53(9):252-260, 341.
	SHEN H L, SU X Q, ZHAO Q L,et al.. Extraction of lodging area of new wheat varieties by unmanned aerial vehicle remote sensing via deep learning [J]. Trans. Chin. Soc. Agric. Mach.,2022,53(9):252-260, 341.

[1]	Ying ZHOU, Jingyong LI, Linxiu DAI, Dicai AO, Ziyi LI, Fan YANG, Junwei GU, Qiang XU, Zhi DOU, Hui GAO. Effect of Melatonin Spraying on Rice Yield Formation and Lodging Resistance Under Rice-Crayfish Coculture Mode [J]. Journal of Agricultural Science and Technology, 2023, 25(9): 34-42.
[2]	Yajun YUAN, Jiaxing FENG, Qifan YANG, Xue BAI, R A J PUSHPA, Dahong BIAN, Yanhong CUI. Lodging-resistance Comparison Among Sumer Maize Varieties with Different Growth Period in North of Huang-Huai-Hai Plain [J]. Journal of Agricultural Science and Technology, 2023, 25(7): 21-28.
[3]	Qiujing CHEN, Zhaodi YANG, Shiyu WANG, Fenggen GUO, Xiaoxue ZHAO, Fan CHEN, Yang FENG. Effects of Plant Growth Retarders on Lodging Resistance and Yield of Quinoa [J]. Journal of Agricultural Science and Technology, 2023, 25(11): 42-48.
[4]	YU Bingxing, WANG Hongfu, WANG Zhenhua, ZHANG Peng, CHENG Kai, YU Aili, YAN Haili, YU Bingjie. Effects of Paclobutrazol on Stalk Characteristics and Lodging Resistance of Foxtail Millet [J]. Journal of Agricultural Science and Technology, 2021, 23(8): 37-44.
[5]	WANG Wenyu, WAN Siyu, ZHANG Xuesong, WANG Xu, LI Jiashuo, ZHENG Guiping*. Effects of Silicon Fertilizer on Lodging Resistance of Kenjing 7 Under Different Tillage Modes [J]. Journal of Agricultural Science and Technology, 2021, 23(4): 145-153.
[6]	ZHANG Ying, MA Yushen, WANG Yamei, LIU Zehang, XU Yuejun, XING Hucheng, JIE Yucheng. Genotype Differences in Cadmium Accumulation Ability of Ramie on Cadmium-polluted Field [J]. Journal of Agricultural Science and Technology, 2021, 23(12): 54-65.
[7]	XIANG Wei, MA Lan, LIU Jiajie, DENG Xin, LONG Chaohai, WEN Qinghua, LIU Chunjie, LYU Jiangnan*. Development Trend of Ramie Mechanized Production Technology Based on Patent Analysis of IncoPat [J]. Journal of Agricultural Science and Technology, 2021, 23(1): 107-118.
[8]	JIANG Mingjin1, LI Min1*, ZHOU Weijia1, LUO Deqiang1, JIANG Xuehai1, JI Guangmei1, LI Lijiang1, XU Wenbo1, QIN Yuhe2. Screening and Evaluation of High-quality Indica Rice Cultivars Suitable for Mechanized Transplanting [J]. Journal of Agricultural Science and Technology, 2020, 22(10): 27-37.
[9]	YANG Anming1, WU Mingliang1,2*, LYU Jiangnan3, XIANG Wei3. Design and Experiment on Chopping System of Forage Ramie Harvester [J]. Journal of Agricultural Science and Technology, 2020, 22(10): 110-118.
[10]	LI Linlin1, HUANG Minsheng1, CUI Guoxian1,2*, BAI Yuchao1,3, LI Xueling1, LIU Nannan1, CUI Dandan1, SU Xiaohui1, WANG Jilong1. Principal Component and Cluster Analysis of the Main Agronomic Characters of Ramie Germplasm [J]. Journal of Agricultural Science and Technology, 2019, 21(3): 34-41.
[11]	XIANG Wei1, MA Lan1, LIU Jiajie1, XIAO Le2, LONG Chaohai1, WEN Qinghua1, LYU Jiangnan1*. Research Progress on Technology and Equipment of Ramie Fibre Stripping and Processing in China [J]. Journal of Agricultural Science and Technology, 2019, 21(11): 59-69.
[12]	YU Miao, CHEN Honggao*, WANG Chaoqun, SHI Wei. Isolation and Characterization of Two Predominant Strains from Ramie Degumming Bacterial Consortium RAMCD407 [J]. , 2013, 15(4): 144-150.
[13]	ZHENG Ke, DUAN Sheng-wen, LIU Zheng-chu, FENG Xiang-yuan, CHENG Li-feng, ZHENG. Breeding of High Efficient Degumming Strain Erwinia carotovora CXJZU-120 for Fiber Crops [J]. , 2011, 13(6): 72-77.