狗牙根叶片相对含水量高光谱反演估算

doi:10.11686/cyxb2024039

摘要/Abstract

摘要：

为了探讨高光谱技术在监测植物受干旱胁迫程度及筛选抗旱材料中的应用，以18个狗牙根基因型为材料，进行为期12 d的自然干旱，测定各材料土壤含水量和叶片相对含水量，同时利用美国SVC HR-768便携式光谱仪，获取不同干旱时间各材料的高光谱相片，以经过高光谱SG平滑和SG平滑+一阶导数结合处理的光谱反射率作为自变量，经Pearson相关性分析和连续投影筛选与叶片相对含水量相关性较好且各材料共有的特征波段，并利用方差膨胀因子检验其共线性，之后通过BP神经网络、支持向量机和随机森林3种机器学习算法建立狗牙根叶片相对含水量的反演模型。结果表明：1）通过连续投影算法共筛选出5个SG平滑+一阶导数特征波段，分别为406、569、706、736、786 nm，与叶片相对含水量的相关性较高（P>0.5），且波段间共线性较弱，可作为抗旱监测的敏感波段。2）以敏感波段为基础建立的随机森林反演模型，决定系数（R²）和均方根误差（RMSE）分别为0.939和8.552，相较于支持向量机和BP神经网络，R²分别提高了5%和8%，表现出最好的预测效果和普适性，其测试集的R²为0.925，RMSE为9.008。研究结果可为未来利用高光谱进行广谱性狗牙根叶片相对含水量无损高精监测提供技术支撑。

关键词: 狗牙根, 不同基因型, 叶片相对含水量, 高光谱技术

Abstract:

Hyperspectral techniques have been widely used to monitor the degree of drought stress and screen for drought-resistant plant materials. In this study， 18 bermudagrass genotypes were subjected to natural drought for 12 days. The soil water content and leaf relative water content of each material were determined， and hyperspectral photos were obtained at different timepoints during the drought treatment using an SVC HR-768 portable spectrometer. The spectral reflectance， which was processed using a combination of hyperspectral Savitzkye Golay smoothing and Savitzkye Golay smoothing+first derivative， was used as the independent variable， and characteristic bands that were well correlated with the relative water content of leaves and， on the basis of Pearson’s correlation and continuous projection analyses， were common among all the materials screened. Then， an inversion model of the relative water content of bermudagrass leaves was established using three machine-learning algorithms： the BP neural network， support vector machine， and random forest algorithms. The main results were as follows： 1） Five Savitzkye Golay smoothing+first derivative characteristic bands were screened out using the continuous projection algorithm； the bands were located at 406， 569， 706， 736， and 786 nm， respectively， and showed high correlations （P>0.5） with the relative water content of bermudagrass leaves. The covariance among the bands was weak， so these bands could be used as sensitive bands for drought monitoring. 2） The coefficient of determination （R² ） and root-mean-square error （RMSE） of the random forest inversion model based on the sensitive wavebands were 0.939 and 8.552， respectively， which were 5% and 8% higher， respectively， than those of the support vector machine and BP neural network models. Thus， the random forest inversion model showed the best prediction effect and universality. The R² of the test set was 0.925， and the RMSE was 9.008. The results of our study provide technical support for the development of a non-destructive and high-precision method to monitor the relative water content of bermudagrass leaves using hyperspectroscopy.

Key words: bermudagrass, different genotypes, leaf relative water content, hyperspectral techniques

喻启坤, 李雯, 汤丽斯, 韩宇, 李培英, 孙宗玖. 狗牙根叶片相对含水量高光谱反演估算[J]. 草业学报, 2024, 33(12): 59-72.

Qi-kun YU, Wen LI, Li-si TANG, Yu HAN, Pei-ying LI, Zong-jiu SUN. Estimation of relative water content in bermudagrass leaves based on hyperspectroscopy[J]. Acta Prataculturae Sinica, 2024, 33(12): 59-72.

图/表 14

参考文献 24

1	Chan Z L， Shi H T， Wang Y P. Response of bermuda grass to abiotic stress. Pratacultural Science， 2013， 30（8）： 1182-1187.
	产祝龙，施海涛，王艳平. 狗牙根抗非生物胁迫的研究进展. 草业科学， 2013， 30（8）： 1182-1187.
2	Abulaiti， Shi D S， Yang G， et al. Preliminary research report on the native Cynodon dactylon in Xinjiang. Journal of Xinjiang Agricultural University， 1998， 21（2）： 124-127.
	阿不来提，石定燧，杨光，等. 新疆野生狗牙根研究初报. 新疆农业大学学报， 1998， 21（2）： 124-127.
3	Qian Y L， Fry J D. Water relations and drought tolerance of four turfgrasses. Journal of the American Society for Horticultural Science， 1997， 122（1）： 129-133.
4	Yang L L， Hua K， Zhang X X. Physiological response and spectral characteristics of tall fescue under different drought stress and CO₂ concentrations. Chinese Journal of Grassland， 2014， 36（4）： 72-78.
	杨璐璐，华开，张学霞. 不同CO₂浓度及干旱胁迫下高羊茅的生理响应和光谱特征. 中国草地学报， 2014， 36（4）： 72-78.
5	Zhao Z J， Shan G L， Duan X H， et al. Study on spectral reflectance and physiological characteristics of three cool-season turfgrass under drought stress. Grassland and Turf， 2016， 36（6）： 23-29.
	赵志军，单贵莲，段新慧，等. 干旱胁迫对3种冷季型草坪草光谱反射率及生理特征的影响. 草原与草坪， 2016， 36（6）： 23-29.
6	Jiang Y， Liu H， Cline V. Correlations of leaf relative water content， canopy temperature， and spectral reflectance in perennial ryegrass under water deficit conditions. HortScience： A Publication of the American Society for Horticultural Science， 2009， 44（2）： 459-462.
7	Chen G， Che W G， Jiang H. Turf hyperspectral analysis techniques and their applications. Yunnan： Yunnan Publishing Group Corporation， 2009： 25-26.
	陈功，车伟光，姜华. 草坪高光谱分析技术及其应用. 云南：云南出版集团公司， 2009： 25-26.
8	Jiang Y， Carrow R N. Broadband spectral reflectance models of turfgrass species and cultivars to drought stress. Crop Science， 2007， 47（4）： 1611-1618.
9	Cen H， He Y. Theory and application of near infrared reflectance spectroscopy in determination of food quality. Trends in Food Science & Technology， 2007， 18（2）： 72-83.
10	Perkins J H， Tenge B， Honigs D E. Resolution enhancement using an approximate-inverse Savitzky-Golay smooth. Spectrochimica Acta Part B： Atomic Spectroscopy， 1988， 43（4/5）： 575-603.
11	Pontes M J C， Santos S R B， Araujo M C U， et al. Classification of distilled alcoholic beverages and verification of adulteration by near infrared spectrometry. Food Research International， 2006， 39（2）： 182-189.
12	Jia Z C， Wang Z J， Li X Y， et al. Marine sediment particle size classification based on the fusion of principal component analysis and continuous projection algorithm. Spectroscopy and Spectral Analysis， 2023， 43（10）： 3075-3080.
	贾宗潮，王子鉴，李雪莹，等. 主成分分析和连续投影融合的海洋沉积物粒度分类研究. 光谱学与光谱分析， 2023， 43（10）： 3075-3080.
13	Gou J， Liu G， He J. Estimation of SPAD values for potato leaves based on least squares support vector machine. Hunan Agricultural Sciences， 2021， 8（11）： 82-86.
	苟静，刘刚，何敬. 基于最小二乘支持向量机的马铃薯叶片SPAD值估算. 湖南农业科学， 2021， 8（11）： 82-86.
14	Wang W D， Chang Q R， Wang Y N. Hyperspectral monitoring of anthocyanins relative content in winter wheat leaves. Journal of Triticeae Crops， 2020， 40（6）： 754-761.
	王伟东，常庆瑞，王玉娜. 冬小麦叶片花青素相对含量高光谱监测. 麦类作物学报， 2020， 40（6）： 754-761.
15	He Y， Peng J Y， Liu F， et al. Critical review of fast detection of crop nutrient and physiological information with spectral and imaging technology. Transactions of the Chinese Society of Agricultural Engineering， 2015， 31（3）： 174-189.
	何勇，彭继宇，刘飞，等. 基于光谱和成像技术的作物养分生理信息快速检测研究进展. 农业工程学报， 2015， 31（3）： 174-189.
16	Niu F P， Li X G， Mamattursun·Eziz， et al. Hyperspectral estimation of soil organic carbon content in the west lakeside oasis of Bosten Lake based on successive projection algorithm. Journal of Zhejiang University （Agriculture & Life Sciences）， 2021， 47（5）： 673-682.
	牛芳鹏，李新国，麦麦提吐尔逊·艾则孜，等. 基于连续投影算法的博斯腾湖西岸湖滨绿洲土壤有机碳含量的高光谱估算. 浙江大学学报（农业与生命科学版）， 2021， 47（5）： 673-682.
17	Guo S， Chang Q R， Zhang Y M， et al. Hyperspectral estimation of maize nitrogen balance index by successive projection combined with SSA-ELM. Acta Agriculturae Boreali-Occidentalis Sinica， 2023， 32（1）： 130-138.
	郭松，常庆瑞，张佑铭，等. 连续投影与SSA-ELM结合的玉米氮平衡指数高光谱估测. 西北农业学报， 2023， 32（1）： 130-138.
18	Bannari A， Khurshid K S， Staenz K， et al. A comparison of hyperspectral chlorophyll indices for wheat crop chlorophyll content estimation using laboratory reflectance measurements. IEEE Transactions on Geoscience and Remote Sensing， 2007， 45（10）： 3063-3074.
19	Katuwal K B， Yang H， Huang B. Evaluation of phenotypic and photosynthetic indices to detect water stress in perennial grass species using hyperspectral， multispectral and chlorophyll fluorescence imaging. Grass Research， 2023， 3（1）： 4-16.
20	Wang P P. Hyperspectral prediction model for moisture content of ramie leaves based on SVR. Changsha： Hunan Agricultural University， 2020.
	汪佩佩. 基于SVR的苎麻叶片含水率高光谱预测模型. 长沙：湖南农业大学， 2020.
21	Guo R， Fu S， Hou M J， et al. Remote sensing retrieval of nature grassland biomass in Menyuan County， Qinghai Province experimental area based on Sentinel-2 data. Acta Prataculturae Sinica， 2023， 32（4）： 15-29.
	郭芮，伏帅，侯蒙京，等. 基于Sentinel-2数据的青海门源县天然草地生物量遥感反演研究. 草业学报， 2023， 32（4）： 15-29.
22	Jiang Y Y， Liu B W， Zhang C J， et al. Multi-variety maize maturity monitoring based on UAV multi-spectral images. Transactions of the Chinese Society of Agricultural Engineering， 2023， 39（20）： 84-91.
	姜友谊，刘博伟，张成健，等. 利用无人机多光谱影像的多品种玉米成熟度监测. 农业工程学报， 2023， 39（20）： 84-91.
23	Yue Y K， Chen J F， Zhao L， et al. Inversion of chlorophyll content in ramie based on UAV multispectral remote sensing. Shandong Agricultural Sciences， 2023， 55（7）： 152-158.
	岳云开，陈建福，赵亮，等. 基于无人机多光谱遥感的苎麻叶绿素含量反演. 山东农业科学， 2023， 55（7）： 152-158.
24	Tao X Y， Zhu Y J， Su X X， et al. Estimation of nitrogen nutrition before flowering stage of winter wheat based on UAV multispectral imagery. Journal of Anhui Science and Technology University， 2023， 37（3）： 50-59.
	陶新宇，朱永基，苏祥祥，等. 基于无人机多光谱影像的冬小麦花前期氮素营养估测. 安徽科技学院学报， 2023， 37（3）： 50-59.

项目 Item	干旱时间 Drought time
项目 Item	0 d	3 d	6 d	9 d	12 d
C136	90.22±5.19Aab	74.41±16.60Aa	35.52±19.86Ba	14.26±14.92BCa	4.91±3.31Ca
C121	97.00±5.19Aa	70.01±10.50Ba	18.20±0.63Cabc	7.31±1.39Da	3.69±0.57Da
C100	93.39±5.74Aab	61.18±21.04Bab	15.56±9.13Cabc	14.05±10.94Ca	3.87±1.14Ca
C99	84.50±0.47Aab	53.91±10.38Bab	20.30±16.70Cabc	10.19±6.63Ca	3.06±0.71Ca
C92	90.15±4.07Aab	60.10±6.02Bab	13.39±5.15Cabc	9.96±3.77CDa	4.34±2.30Da
C87	84.18±13.52Aab	56.39±14.12Bab	9.88±5.34Cbc	7.64±0.37Ca	4.15±1.17Ca
C84	80.97±11.63Ab	65.00±18.82Aa	13.01±8.24Bbc	7.24±0.79Ba	3.95±0.97Ba
C76	83.23±3.86Aab	48.20±14.36Bab	10.72±7.67Cbc	9.87±2.38Ca	4.09±0.81Ca
C75	81.09±9.36Ab	57.29±13.27Aab	26.96±26.18Babc	7.07±3.07Ba	4.29±2.07Ba
C72	84.67±4.54Aab	60.56±19.65Bab	6.35±1.18Cc	8.90±3.47Ca	3.82±1.62Ca
C63	84.74±6.82Aab	63.32±10.07Bab	15.54±8.09Cabc	7.49±0.97Ca	4.27±3.40Ca
C40	90.85±9.85Aab	48.02±13.85Bab	13.48±6.11Cabc	6.30±0.16Ca	3.16±0.83Ca
C28	92.89±6.16Aab	59.30±18.39Bab	20.62±19.99Cabc	12.10±8.29Ca	3.65±1.01Ca
C22	91.87±3.03Aab	68.87±10.92Ba	31.19±15.50Cab	9.44±4.46Da	3.98±1.38Da
X1	96.11±6.74Aa	65.50±27.13Ba	13.59±3.27Cabc	6.86±1.99Ca	4.31±1.09Ca
X2	94.66±8.67Aab	63.61±6.53Ba	8.10±1.78Cc	7.39±1.21Ca	5.21±0.30Ca
KS	96.87±5.42Aa	46.30±11.16Bab	12.44±7.65Cbc	6.88±0.32Ca	4.89±1.29Ca
PG	88.93±1.23Aab	33.79±12.71Bb	6.44±2.01Cc	6.78±1.11Ca	3.44±0.33Ca
最大值Maximun	97.00	74.41	35.52	14.26	5.21
最小值Minimum	80.97	33.79	6.35	6.30	3.06
平均值Mean	89.24	58.65	16.18	8.87	4.06
标准差Standard deviation	5.41	9.85	8.15	2.47	0.57
变异系数Variable coefficient	6.06	16.80	50.36	27.80	14.10

项目 Item	干旱时间 Drought time
项目 Item	0 d	3 d	6 d	9 d	12 d
C136	90.22±5.19Aab	74.41±16.60Aa	35.52±19.86Ba	14.26±14.92BCa	4.91±3.31Ca
C121	97.00±5.19Aa	70.01±10.50Ba	18.20±0.63Cabc	7.31±1.39Da	3.69±0.57Da
C100	93.39±5.74Aab	61.18±21.04Bab	15.56±9.13Cabc	14.05±10.94Ca	3.87±1.14Ca
C99	84.50±0.47Aab	53.91±10.38Bab	20.30±16.70Cabc	10.19±6.63Ca	3.06±0.71Ca
C92	90.15±4.07Aab	60.10±6.02Bab	13.39±5.15Cabc	9.96±3.77CDa	4.34±2.30Da
C87	84.18±13.52Aab	56.39±14.12Bab	9.88±5.34Cbc	7.64±0.37Ca	4.15±1.17Ca
C84	80.97±11.63Ab	65.00±18.82Aa	13.01±8.24Bbc	7.24±0.79Ba	3.95±0.97Ba
C76	83.23±3.86Aab	48.20±14.36Bab	10.72±7.67Cbc	9.87±2.38Ca	4.09±0.81Ca
C75	81.09±9.36Ab	57.29±13.27Aab	26.96±26.18Babc	7.07±3.07Ba	4.29±2.07Ba
C72	84.67±4.54Aab	60.56±19.65Bab	6.35±1.18Cc	8.90±3.47Ca	3.82±1.62Ca
C63	84.74±6.82Aab	63.32±10.07Bab	15.54±8.09Cabc	7.49±0.97Ca	4.27±3.40Ca
C40	90.85±9.85Aab	48.02±13.85Bab	13.48±6.11Cabc	6.30±0.16Ca	3.16±0.83Ca
C28	92.89±6.16Aab	59.30±18.39Bab	20.62±19.99Cabc	12.10±8.29Ca	3.65±1.01Ca
C22	91.87±3.03Aab	68.87±10.92Ba	31.19±15.50Cab	9.44±4.46Da	3.98±1.38Da
X1	96.11±6.74Aa	65.50±27.13Ba	13.59±3.27Cabc	6.86±1.99Ca	4.31±1.09Ca
X2	94.66±8.67Aab	63.61±6.53Ba	8.10±1.78Cc	7.39±1.21Ca	5.21±0.30Ca
KS	96.87±5.42Aa	46.30±11.16Bab	12.44±7.65Cbc	6.88±0.32Ca	4.89±1.29Ca
PG	88.93±1.23Aab	33.79±12.71Bb	6.44±2.01Cc	6.78±1.11Ca	3.44±0.33Ca
最大值Maximun	97.00	74.41	35.52	14.26	5.21
最小值Minimum	80.97	33.79	6.35	6.30	3.06
平均值Mean	89.24	58.65	16.18	8.87	4.06
标准差Standard deviation	5.41	9.85	8.15	2.47	0.57
变异系数Variable coefficient	6.06	16.80	50.36	27.80	14.10

编号 Code	干旱时间 Drought time
编号 Code	0 d	3 d	6 d	9 d	12 d
C136	94.57±4.72abc	84.11±14.42abc	68.36±2.93bc	49.43±7.59a	26.00±3.84a
C121	99.51±0.62ab	86.01±1.19abc	67.51±2.43bc	18.48±5.68bcdef	15.08±2.70bcd
C100	95.13±4.60abc	87.80±10.46abc	73.90±1.56ab	21.77±2.52bcde	14.99±2.47bcd
C99	95.31±5.90abc	77.35±8.19abc	58.79±7.00def	24.37±11.38bcd	15.11±4.26bcd
C92	99.87±0.00a	92.13±13.39ab	68.59±3.49bc	25.00±5.37bc	14.21±1.06bcde
C87	99.06±1.50abc	89.73±8.79abc	57.81±1.97efg	16.82±7.34cdefg	13.17±1.05cdefg
C84	96.82±4.22abc	95.74±7.14ab	66.48±1.48bcd	24.89±0.74bc	12.43±1.48cdefgh
C76	96.30±3.15abc	84.29±0.82abc	57.00±3.11efg	13.19±0.90efg	10.36±1.22efghi
C75	94.30±9.66abc	87.30±10.93abc	64.89±6.77cde	25.22±2.01bc	16.38±3.25bc
C72	95.47±4.21abc	90.10±8.62abc	57.89±6.07efg	14.94±5.36efg	9.55±2.11fghi
C63	98.54±2.29abc	99.87±0.00a	71.88±8.49abc	25.68±3.28bc	14.81±3.62bcd
C40	99.87±0.01a	95.32±7.86ab	49.89±4.59g	8.92±0.51g	8.23±0.23hi
C28	98.60±1.31abc	99.88±0.00a	68.14±5.19bc	15.01±0.26efg	13.63±0.53bcdef
C22	93.06±7.10abc	88.78±9.66abc	76.99±5.22a	26.07±1.88b	17.96±2.79b
X1	89.91±7.64bc	78.36±37.21abc	58.92±2.10def	13.65±2.23efg	7.77±0.42i
X2	89.49±6.09c	75.41±3.81bc	67.26±3.09bcd	18.03±3.70bcdef	9.07±1.25ghi
KS	91.77±7.06abc	92.17±13.29ab	51.14±6.51fg	15.82±1.95defg	8.47±2.44hi
PG	91.30±4.97abc	67.87±1.10c	13.14±1.80h	11.40±1.93fg	11.04±0.38defghi
最大值Maximun	99.87	99.88	76.99	49.43	26.00
最小值Minimum	89.49	67.87	13.14	8.92	7.77
平均值Mean	95.49	87.34	61.03	20.48	13.24
标准差Standard deviation	3.39	8.52	14.11	9.07	4.41
变异系数Variable coefficient	3.55	9.75	23.13	44.26	33.34

编号 Code	干旱时间 Drought time
编号 Code	0 d	3 d	6 d	9 d	12 d
C136	94.57±4.72abc	84.11±14.42abc	68.36±2.93bc	49.43±7.59a	26.00±3.84a
C121	99.51±0.62ab	86.01±1.19abc	67.51±2.43bc	18.48±5.68bcdef	15.08±2.70bcd
C100	95.13±4.60abc	87.80±10.46abc	73.90±1.56ab	21.77±2.52bcde	14.99±2.47bcd
C99	95.31±5.90abc	77.35±8.19abc	58.79±7.00def	24.37±11.38bcd	15.11±4.26bcd
C92	99.87±0.00a	92.13±13.39ab	68.59±3.49bc	25.00±5.37bc	14.21±1.06bcde
C87	99.06±1.50abc	89.73±8.79abc	57.81±1.97efg	16.82±7.34cdefg	13.17±1.05cdefg
C84	96.82±4.22abc	95.74±7.14ab	66.48±1.48bcd	24.89±0.74bc	12.43±1.48cdefgh
C76	96.30±3.15abc	84.29±0.82abc	57.00±3.11efg	13.19±0.90efg	10.36±1.22efghi
C75	94.30±9.66abc	87.30±10.93abc	64.89±6.77cde	25.22±2.01bc	16.38±3.25bc
C72	95.47±4.21abc	90.10±8.62abc	57.89±6.07efg	14.94±5.36efg	9.55±2.11fghi
C63	98.54±2.29abc	99.87±0.00a	71.88±8.49abc	25.68±3.28bc	14.81±3.62bcd
C40	99.87±0.01a	95.32±7.86ab	49.89±4.59g	8.92±0.51g	8.23±0.23hi
C28	98.60±1.31abc	99.88±0.00a	68.14±5.19bc	15.01±0.26efg	13.63±0.53bcdef
C22	93.06±7.10abc	88.78±9.66abc	76.99±5.22a	26.07±1.88b	17.96±2.79b
X1	89.91±7.64bc	78.36±37.21abc	58.92±2.10def	13.65±2.23efg	7.77±0.42i
X2	89.49±6.09c	75.41±3.81bc	67.26±3.09bcd	18.03±3.70bcdef	9.07±1.25ghi
KS	91.77±7.06abc	92.17±13.29ab	51.14±6.51fg	15.82±1.95defg	8.47±2.44hi
PG	91.30±4.97abc	67.87±1.10c	13.14±1.80h	11.40±1.93fg	11.04±0.38defghi
最大值Maximun	99.87	99.88	76.99	49.43	26.00
最小值Minimum	89.49	67.87	13.14	8.92	7.77
平均值Mean	95.49	87.34	61.03	20.48	13.24
标准差Standard deviation	3.39	8.52	14.11	9.07	4.41
变异系数Variable coefficient	3.55	9.75	23.13	44.26	33.34

特征波段 Characteristic band （nm）	方差膨胀因子 Variance inflation factor （VIF）
476	6.977
522	5.238
556	11.942
681	3.309
739	6.953
778	8.625
812	6.488