[1]李明,陶光林*,廖华刚.基于冠层高光谱数据的油茶炭疽病病情指数估测[J].江苏林业科技,2023,50(04):25-29.[doi:10.3969/j.issn.1001-7380.2023.04.005]
　Li Ming,Tao Guanglin*,Liao Huagang.Estimation of anthracnose disease index in Camellia oleifera based on canopy hyperspectral data[J].Journal of Jiangsu Forestry Science &Technology,2023,50(04):25-29.[doi:10.3969/j.issn.1001-7380.2023.04.005]
点击复制

基于冠层高光谱数据的油茶炭疽病病情指数估测()
分享到：

《江苏林业科技》[ISSN:1001-7380/CN:32-1236/S]

卷:
第50卷
期数:
2023年04期
页码:
25-29
栏目:
试验研究
出版日期:
2023-08-31

文章信息/Info

Title:
Estimation of anthracnose disease index in Camellia oleifera based on canopy hyperspectral data
文章编号:
1001-7380（2023）04-0025-05
作者:
李明; 陶光林*; 廖华刚
黔东南州林业科学研究所，贵州 凯里 556000
Author(s):
Li Ming; Tao Guanglin*;  Liao Huagang
Qiandongnan Institute of Forestry，Kaili 556000， China
关键词:
油茶; 炭疽病; 连续投影算法; 竞争性自适应重加权算法
Keywords:
Camellia oleifera; Anthracnose disease; Successive projections algorithm; Competency adaptive reweighting sampling
分类号:
Q433.4；S763.13；S794.4
DOI:
10.3969/j.issn.1001-7380.2023.04.005
文献标志码:
A
摘要:
利用一阶导数、S-G平滑及多元散射校正等光谱预处理方法对油茶炭疽病危害下原始光谱进行预处理后，采用SPXY样本划分法将65个样本按7∶3将样本划分为45个校正集和20个预测集，再对不同预处理的光谱数据建立油茶炭疽病病情指数偏最小二乘回归模型。结果显示，在多种预处理方法中，S-G平滑预处理效果最好。通过连续投影算法(SPA)以及竞争性自适应重加权采样算法(CARS)，从S-G平滑预处理的光谱中提取特征波长，进而构建基于偏最小二乘回归(PLSR)的油茶炭疽病病情指数的估测模型。试验结果发现，基于SPA所提特征波长建立的SPA-PLSR模型预测集R2p和预测均方根偏差(RMSEP)分别为0.700和0.072，基于CARS所提特征波长建立的CARS-PLSR模型预测集R2p和RMSEP分别为0.834和0.053；CARS-PLSR预测模型总体上要优于SPA-PLSR模型，可实现油茶炭疽病病情指数的估测。
Abstract:
The original spectra of camellia oleifera affected by anthracnose disease were firstly preprocessed by first-order derivative, S-G smoothing, and multiplicative scattering correction. The SPXY (sample set partitioning based on joint x-y distance) method was then employed to divide 65 samples into 45 calibration sets and 20 prediction sets at a ratio of 7∶3. Subsequently, partial least squares regression (PLSR) models for estimating the anthracnose disease index were established based on different preprocessed spectral data. The result showed that among various preprocessing methods, the S-G smoothing worked best. Feature wavelengths were extracted from the S-G smoothed spectra using successive projections algorithm (SPA) and competitive adaptive reweighting sampling algorithm (CARS). The PLSR models based on SPA-selected feature wavelengths (SPA-PLSR) and CARS-selected ones (CARS-PLSR) were constructed to predict the anthracnose disease index. Experimental results demonstrated that the SPA-PLSR model achieved a prediction set R2p of 0.700 and RMSEP of 0.072, while the CARS-PLSR model got R2p of 0.834 and RMSEP of 0.053.Overall, the CARS-PLSR model outperformed the SPA-PLSR model, and provided a feasible approach for estimating anthracnose disease index in Camellia oleifera.

参考文献/References:

[1]秦绍钊,张柱亭,王洪.油茶炭疽病防治技术探析[J].南方农机,2020,51(8):10.
[2]赵丹阳,秦长生.油茶病虫害诊断与防治原色生态图谱[M].广州:广东科技出版社,2018.
[3]侯冰茹,刘鹏辉,张洋,等.基于马铃薯叶片光纤光谱信息的晚疫病患病程度预测[J].光谱学与光谱分析,2022,42(5):1426-1432.
[4]邓小玲,曾国亮,朱梓豪,等.基于无人机高光谱遥感的柑橘患病植株分类与特征波段提取[J].华南农业大学学报,2020,41(6):100-108.
[5]冯子恒,李晓,段剑钊,等.基于特征波段选择和机器学习的小麦白粉病高光谱遥感监测[J].作物学报,2022,48(9):2300-2314.
[6]国家林业局.油茶主要有害生物综合防技术规程[Z].2016-07-27.
[7]刘君昂,潘华平,伍南,等.油茶主要病害空间分布格局规律的研究[J].中国森林病虫,2010,29(5):7-10.
[8]GALVO R K H, ARA JO M C U, FRAGOSO W D, et al. A variable elimination method to improve the parsimony of MLR models using the successive projections algorithm[J].Chemometrics and Intelligent Laboratory Systems, 2008, 92(1):83-91.
[9]LI H D, LIANG Y Z, XU Q S,et al. Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration[J]. Analytica Chimica Acta,2009,648(1):77-84.
[10]黄清,薛河儒,刘江平,等.基于蚁群-遗传算法的光谱选择方法与应用[J].光谱学与光谱分析,2022,42(7):2262-2268.
[11]QU F F, REN D, WANG J H, et al. An ensemble successive project algorithm for liquor detection using near infrared sensor [J].Sensors,2016,16(1):89.

相似文献/References:

[1]董筱昀,黄利斌,吕运舟,等.江苏油茶种质资源调查与分析[J].江苏林业科技,2017,44(01):34.[doi:10.3969/j.issn.1001-7380.2017.01.007]
　DONG Xiao-yun,HUANG Li-bin,LYU Yun-zhou,et al.Survey andanalysis on tea-oil Camellia resource in Jiangsu Province[J].Journal of Jiangsu Forestry Science &Technology,2017,44(04):34.[doi:10.3969/j.issn.1001-7380.2017.01.007]
[2]郭小华,钟子龙,严志伟,等.低产老油茶林品种置换技术研究[J].江苏林业科技,2020,47(01):46.[doi:10.3969/j.issn.1001-7380.2020.01.009]
[3]毛云飞,石兴华,谢爱香,等.林下套种对老龄油茶产量及林地土壤养分的影响[J].江苏林业科技,2020,47(04):32.[doi:10.3969/j.issn.1001-7380.2020.04.008]
[4]王晓艳.不同菌肥对油茶叶内源激素及氮磷钾含量和林下土壤理化性质的影响[J].江苏林业科技,2021,48(05):28.[doi:10.3969/j.issn.1001-7380.2021.05.004]
　Wang Xiaoyan.Effects of different bacterial fertilizers on soil properties，leaf endogenous hormones and N，P，K contents in Camellia oleifera Abel.[J].Journal of Jiangsu Forestry Science &Technology,2021,48(04):28.[doi:10.3969/j.issn.1001-7380.2021.05.004]
[5]李明,胡岑龙*,陶光林.基于不同预处理的油茶叶片SPAD值高光谱遥感估算研究[J].江苏林业科技,2022,49(02):1.[doi:10.3969/j.issn.1001-7380.2022.02.001]

备注/Memo

备注/Memo:
收稿日期：2023-05-10;修回日期：2023-06-14
基金项目：贵州省林业局青年人才基金项目“基于植物高光谱特征的油茶主要病虫害监测及诊断模型研究”（黔林科合J字〔2020〕06号）
作者简介：李明（1986- ），男，贵州镇远人，工程师，硕士。从事林业3S技术研究。
*通信作者：陶光林（1971- ），男，高级工程师。从事野生植物引种栽培技术研究。
更新日期/Last Update: 2023-11-15