基于BiPLS-CARS-PLS的哈密瓜冠层叶片SPAD值反演建模

doi:10.3969/j.issn.1004-1524.2022.10.23

浙江农业学报 ›› 2022, Vol. 34 ›› Issue (10): 2286-2295.DOI: 10.3969/j.issn.1004-1524.2022.10.23

基于BiPLS-CARS-PLS的哈密瓜冠层叶片SPAD值反演建模

郭阳^a(), 郭俊先^a, 史勇^a^,^*(), 李雪莲^a, 黄华^b

^a 机电工程学院新疆农业大学,新疆乌鲁木齐 830052
^b 数理学院新疆农业大学,新疆乌鲁木齐 830052

收稿日期:2021-05-25 出版日期:2022-10-25 发布日期:2022-10-26
通讯作者: 史勇
作者简介:*史勇,E-mail: 280974136@qq.com
郭阳(1995—),男,辽宁沈阳人,硕士研究生,研究方向为农产品无损检测。E-mail: 2744103108@qq.com
基金资助:
新疆维吾尔自治区教育厅自然科学重点项目(XJEDU2020I009);国家自然科学基金(61367001)

SPAD inversion model of cantaloupe canopy leaf based on BiPLS-CARS-PLS

GUO Yang^a(), GUO Junxian^a, SHI Yong^a^,^*(), LI Xuelian^a, HUANG Hua^b

^a College of Electrical and Mechanical Engineering, Xinjiang Agricultural University, Urumqi 830052, China
^b College of Mathematics and Physics, Xinjiang Agricultural University, Urumqi 830052, China

Received:2021-05-25 Online:2022-10-25 Published:2022-10-26
Contact: SHI Yong

摘要/Abstract

摘要：

利用光谱技术对大田哈密瓜冠层叶片叶绿素含量定量估测,可为田间水肥调控以及田间管理提供理论依据。本实验在剔除噪音后的378 nm到1 115 nm光谱的基础上采用多元散射校正、标准正态变量相交、标准化、Savitzky-Golay卷积平滑法、归一化、移动平均平滑等方法对原始光谱数据进行预处理,然后采用特征区间选择与特征波长选择相结合的方法实现数据降维和简化模型,并建立偏最小二乘和极限学习机的回归模型。结果表明,多元散射校正预处理效果最佳,在此基础上,利用反向区间偏最小二乘法(BiPLS)和竞争性自适应重加权采样算法(CARS)相结合共筛选出13个特征波长,将其作为模型的输入变量,由偏最小二乘法(PLS)建立的模型效果最优,其预测集的相关系数R_p和均方根误差RMSEP分别为0.942 4与1.006 2。因此,采用BiPLS与 CARS结合PLS建立的光谱定量分析模型,可实现对哈密瓜冠层叶片叶绿素含量的定量估测。

关键词: 哈密瓜冠层, SPAD值, 特征区间选择, 特征波长选择, 偏最小二乘法, 估算模型

Abstract:

The quantitative estimation of chlorophyll content in canopy leaves of Hami melon by spectral technology can provide a theoretical basis for water and fertilizer regulation and field management. On the basis of the original spectrum, this experiment used multivariate scattering correction, standard normal variable intersection, standardization, Savitzky-Golay convolution smoothing, normalization and moving average smoothing to preprocess the original spectrum data, and then combined the feature interval selection with the feature wavelength selection to achieve the purpose of data dimension reduction and model simplification. The regression models of partial least squares and extreme learning machine were established. The results showed that the preprocessing effect of multiple scattering correction was the best. On this basis, 13 characteristic wavelengths were selected by combining the inverse interval partial least squares (BiPLS) and the competitive adaptive reweighting sampling (CARS) algorithm, which were used as the input variables of the model. The model established by partial least squares (PLS) had the best effect. The correlation coefficient R_p and root mean square error (RMSEP) of the prediction set were 0.942 4 and 1.006 2, respectively. Therefore, the spectral quantitative analysis model established by BiPLS and CARS combined with PLS could realize the quantitative estimation of chlorophyll content in Hami melon canopy leaves.

Key words: canopy of Hami melon, SPAD value, feature interval selection, characteristic wavelength selection, partial least squares, estimation model

中图分类号:

S652.1

郭阳, 郭俊先, 史勇, 李雪莲, 黄华. 基于BiPLS-CARS-PLS的哈密瓜冠层叶片SPAD值反演建模[J]. 浙江农业学报, 2022, 34(10): 2286-2295.

GUO Yang, GUO Junxian, SHI Yong, LI Xuelian, HUANG Hua. SPAD inversion model of cantaloupe canopy leaf based on BiPLS-CARS-PLS[J]. Acta Agriculturae Zhejiangensis, 2022, 34(10): 2286-2295.

图/表 11

参考文献 21

[1]	丁永军, 李民赞, 郑立华, 等. 基于近红外光谱小波变换的温室番茄叶绿素含量预测[J]. 光谱学与光谱分析, 2011, 31(11): 2936-2939.
	DING Y J, LI M Z, ZHENG L H, et al. Prediction of chlorophyll content of greenhouse tomato using wavelet transform combined with NIR spectra[J]. Spectroscopy and Spectral Analysis, 2011, 31(11): 2936-2939. (in Chinese with English abstract)
[2]	MARENCO R A, ANTEZANA-VERA S A, NASCIMENTO H C S. Relationship between specific leaf area, leaf thickness, leaf water content and SPAD-502 readings in six Amazonian tree species[J]. Photosynthetica, 2009, 47(2): 184-190. DOI URL
[3]	张珮, 王银红, 李高阳, 等. 基于近红外光谱对桃果实低温贮藏品质的定量检测[J]. 中国食品学报, 2021, 21(5): 355-362.
	ZHANG P, WANG Y H, LI G Y, et al. Quantitative detection of the quality of peach fruit during low temperature storage based on near infrared spectroscopy[J]. Journal of Chinese Institute of Food Science and Technology, 2021, 21(5): 355-362. (in Chinese with English abstract)
[4]	吴迪, 黄凌霞, 何勇, 等. 作物和杂草叶片的可见-近红外反射光谱特性[J]. 光学学报, 2008, 28(8): 1618-1622.
	WU D, HUANG L X, HE Y, et al. Visible-near infrared reflection spectroscopy for crop-weed discrimination[J]. Acta Optica Sinica, 2008, 28(8): 1618-1622. (in Chinese with English abstract) DOI URL
[5]	李红, 张凯, 陈超, 等. 基于高光谱成像技术的生菜冠层含水率检测[J]. 农业机械学报, 2021, 52(2): 211-217.
	LI H, ZHANG K, CHEN C, et al. Detection of moisture content in lettuce canopy based on hyperspectral imaging technique[J]. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(2): 211-217. (in Chinese with English abstract)
[6]	尚静, 孟庆龙, 张艳, 等. 紫外/可见光谱技术无损检测李子可溶性固形物含量[J]. 食品工业科技, 2020, 41(3): 228-231.
	SHANG J, MENG Q L, ZHANG Y, et al. Nondestructive detection of soluble solids content of plums based on UV/VIS spectroscopy technology[J]. Science and Technology of Food Industry, 2020, 41(3): 228-231. (in Chinese with English abstract)
[7]	闫明壮, 王浩云, 吴媛媛, 等. 基于光谱与纹理特征融合的绿萝叶绿素含量检测[J]. 南京农业大学学报, 2021, 44(3): 568-575.
	YAN M Z, WANG H Y, WU Y Y, et al. Detection of chlorophyll content of Epipremnum aureum based on fusion of spectrum and texture features[J]. Journal of Nanjing Agricultural University, 2021, 44(3): 568-575. (in Chinese with English abstract)
[8]	刘宁, 邢子正, 乔浪, 等. 基于模型集群的马铃薯叶绿素检测光谱变量筛选讨论[J]. 光谱学与光谱分析, 2020, 40(7): 2259-2266.
	LIU N, XING Z Z, QIAO L, et al. Discussion on spectral variables selection of potato chlorophyll using model population analysis[J]. Spectroscopy and Spectral Analysis, 2020, 40(7): 2259-2266. (in Chinese with English abstract)
[9]	李雪, 杨瑞楠, 原喆, 等. 油菜籽叶绿素含量近红外光谱快速检测[J]. 中国油料作物学报, 2019, 41(1): 126-129.
	LI X, YANG R N, YUAN Z, et al. Rapid detection of chlorophyll content in rapeseed based on near infrared spectroscopy[J]. Chinese Journal of Oil Crop Sciences, 2019, 41(1): 126-129. (in Chinese with English abstract)
[10]	吴文强, 常庆瑞, 陈涛, 等. 基于PCA-BP神经网络算法桃树叶片SPAD值高光谱估算[J]. 西北林学院学报, 2019, 34(5): 134-140.
	WU W Q, CHANG Q R, CHEN T, et al. Hyperspectral estimation of peach leaf SPAD value based on PCA-BP neural network algorithm[J]. Journal of Northwest Forestry University, 2019, 34(5): 134-140. (in Chinese with English abstract)
[11]	陈晓, 李修华, 王策, 等. 基于反射光谱全波段与双波段甘蔗叶片叶绿素预测模型比较[J]. 中国农业大学学报, 2018, 23(8): 118-124.
	CHEN X, LI X H, WANG C, et al. Comparison of sugarcane leaf chlorophyll prediction models based on full and dual reflected spectral bands[J]. Journal of China Agricultural University, 2018, 23(8): 118-124. (in Chinese with English abstract)
[12]	康丽, 高睿, 孔庆明, 等. 水稻叶片SPAD值高光谱成像估测[J]. 东北农业大学学报, 2020, 51(10): 89-96.
	KANG L, GAO R, KONG Q M, et al. Estimation of SPAD value of rice leaves based on hyperspectral image[J]. Journal of Northeast Agricultural University, 2020, 51(10): 89-96. (in Chinese with English abstract)
[13]	王璐, 李伟凯, 关海鸥, 等. 基于多维度光谱特征波长提取大豆冠层SPAD值估算模型[J]. 江苏大学学报(自然科学版), 2020, 41(3): 334-338.
	WANG L, LI W K, GUAN H O et al. Estimation model of SPAD value for soybean canopy based on multi-dimensional spectral characteristic wavelength extraction[J]. Journal of Jiangsu University (Natural Science Edition), 2020, 41(3): 334-338. (in Chinese with English abstract)
[14]	苗雪雪, 苗莹, 龚浩如, 等. 不同偏最小二乘法的近红外光谱技术测定大米中水分的研究[J]. 分析科学学报, 2019, 35(5): 643-649.
	MIAO X X, MIAO Y, GONG H R, et al. Determination of moisture content in rice by near infrared spectroscopy with different partial least squares[J]. Journal of Analytical Science, 2019, 35(5): 643-649. (in Chinese with English abstract)
[15]	蒋雪松, 周宏平. 基于遗传算法的农产品品质无损检测研究进展[J]. 江苏农业科学, 2013, 41(12): 17-20.
	JIANG X S, ZHOU H P. Research Progress of non-destructive testing of agricultural product quality based on genetic algorithm[J]. Jiangsu Agricultural Sciences, 2013, 41(12): 17-20. (in Chinese)
[16]	唐海涛, 孟祥添, 苏循新, 等. 基于CARS算法的不同类型土壤有机质高光谱预测[J]. 农业工程学报, 2021, 37(2): 105-113.
	TANG H T, MENG X T, SU X X, et al. Hyperspectral prediction on soil organic matter of different types using CARS algorithm[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(2): 105-113. (in Chinese with English abstract)
[17]	王怡淼, 朱金林, 张慧, 等. 基于MC-UVE、GA算法及因子分析对葡萄酒酒精度近红外定量模型的优化研究[J]. 发光学报, 2018, 39(9): 1310-1316.
	WANG Y M, ZHU J L, ZHANG H, et al. Optimization of near infrared quantitative model for wine alcohol content based on MC-UVE, GA algorithm and factor analysis[J]. Chinese Journal of Luminescence, 2018, 39(9): 1310-1316. (in Chinese with English abstract) DOI URL
[18]	刘爽, 于海业, 朴兆佳, 等. 基于高光谱技术反演大豆生理信息的特征波长提取方法研究[J]. 光谱学与光谱分析, 2020, 40(11): 3542-3548.
	LIU S, YU H Y, PIAO Z J, et al. Study on extracting characteristic wavelength of soybean physiological information based on hyperspectral technique[J]. Spectroscopy and Spectral Analysis, 2020, 40(11): 3542-3548. (in Chinese with English abstract)
[19]	孟庆龙, 尚静, 黄人帅, 等. 基于主成分回归的苹果可溶性固形物含量预测模型[J]. 保鲜与加工, 2020, 20(5): 185-189.
	MENG Q L, SHANG J, HUANG R S, et al. Prediction model for soluble solids content of apples based on principal component regression[J]. Storage and Process, 2020, 20(5): 185-189. (in Chinese with English abstract)
[20]	朱哲燕, 刘飞, 张初, 等. 基于中红外光谱技术的香菇蛋白质含量测定[J]. 光谱学与光谱分析, 2014, 34(7): 1844-1848.
	ZHU Z Y, LIU F, ZHANG C, et al. Protein content determination of shiitake mushroom(Lentinus edodes)using mid-infrared spectroscopy and chemometeics[J]. Spectroscopy and Spectral Analysis, 2014, 34(7): 1844-1848. (in Chinese with English abstract)
[21]	何勇, 刘飞, 李晓丽. 光谱及成像技术在农业中的应用[M]. 北京: 科学出版社, 2016: 97-98.

样本集 Sample set	样本数 Sample size	SPAD
样本集 Sample set	样本数 Sample size	平均值 Mean value	最大值 Maximum value	最小值 Minimum value
校正集	75	54.0	63.5	40.5
Calibration set
预测集	25	51.5	61.1	44.5
Prediction set

样本集 Sample set	样本数 Sample size	SPAD
样本集 Sample set	样本数 Sample size	平均值 Mean value	最大值 Maximum value	最小值 Minimum value
校正集	75	54.0	63.5	40.5
Calibration set
预测集	25	51.5	61.1	44.5
Prediction set

光谱预处理 Spectral pretreatment	PC	R_C	RMSEC	R_P	RMSEP	RPD
Origina	8	0.821 5	1.281 6	0.563 5	1.477 8	1.552 1
Autoscales	5	0.752 1	1.398 8	0.740 5	1.354 4	1.974 1
SNVT	5	0.742 8	1.395 7	0.710 5	1.357 8	1.963 2
Savitzky-Golay	8	0.634 3	2.175 7	0.613 5	2.423 5	2.075 6
MSC	5	0.743 6	1.419 2	0.799 6	1.214 4	2.233 9
MA	8	0.821 2	1.287 8	0.563 9	1.477 2	1.552 8
Normalize	12	0.804 5	1.312 4	0.432 6	1.824 5	1.312 4

光谱预处理 Spectral pretreatment	PC	R_C	RMSEC	R_P	RMSEP	RPD
Origina	8	0.821 5	1.281 6	0.563 5	1.477 8	1.552 1
Autoscales	5	0.752 1	1.398 8	0.740 5	1.354 4	1.974 1
SNVT	5	0.742 8	1.395 7	0.710 5	1.357 8	1.963 2
Savitzky-Golay	8	0.634 3	2.175 7	0.613 5	2.423 5	2.075 6
MSC	5	0.743 6	1.419 2	0.799 6	1.214 4	2.233 9
MA	8	0.821 2	1.287 8	0.563 9	1.477 2	1.552 8
Normalize	12	0.804 5	1.312 4	0.432 6	1.824 5	1.312 4

区间总数 Interval total	入选区间总数 Total number of selected intervals	RMSE	入选变量数 Number of variables selected
10	3	1.397 1	499
11	2	1.369 3	303
12	5	1.388 8	693
13	2	1.359 7	256
14	2	1.365 5	238
15	2	1.378 8	221
16	3	1.306 5	312
17	5	1.388 5	489
18	4	1.373 6	370
19	6	1.366 4	438
20	2	1.390 0	166
21	6	1.337 2	475
22	3	1.359 4	227
23	6	1.327 5	433
24	4	1.314 4	278
25	2	1.322 1	133

基于BiPLS-CARS-PLS的哈密瓜冠层叶片SPAD值反演建模

SPAD inversion model of cantaloupe canopy leaf based on BiPLS-CARS-PLS

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 21

相关文章 4

编辑推荐

Metrics

本文评价

序号 Serial number	剔除区间 Culling interval	RMSE	变量个数 Number of variables
16	11	1.448 3	1 664
15	1	1.444 9	1 560
14	12	1.438 3	1 456
13	9	1.431 3	1 352
12	2	1.427 1	1 248
11	3	1.422 6	1 144
10	6	1.411 8	1 040
9	5	1.404 3	936
8	8	1.381 9	832
7	15	1.364 4	728
6	14	1.358 3	624
5	7	1.338 9	520
4	10	1.334 9	416
3	13	1.306 5	312
2	4	1.349 9	208
1	16	1.478 2	104

处理方法 Processing method	变量数量 Number of variables	校正集Calibration set		预测集Prediction set
处理方法 Processing method	变量数量 Number of variables	R_c	RMSEC	R_p	RMSEP
BiPLS	312	0.879 7	1.328 0	0.922 9	1.135 2
BiPLS-SPA	8	0.865 6	1.382 9	0.901 8	1.247 8
BiPLS-GA	28	0.895 8	1.274 5	0.948 5	0.904 2
BiPLS-CARS	13	0.920 8	1.085 1	0.942 4	1.006 2
BiPLS-MCUVE	72	0.881 9	1.317 5	0.909 8	1.220 6

处理方法 Processing method	变量个数 Number of variables	校正集Calibration set		预测集Prediction set
处理方法 Processing method	变量个数 Number of variables	R_c	RMSEC	R_p	RMSEP
BiPLS	312	0.801 7	1.547 1	0.782 1	1.898 5
BiPLS-SPA	8	0.832 8	1.276 1	0.813 8	1.562 1
BiPLS-GA	28	0.819 9	1.482 5	0.800 9	1.629 2
BiPLS-CARS	13	0.860 8	1.084 6	0.871 7	1.517 3
BiPLS-MCUVE	72	0.806 1	1.510 2	0.799 5	1.081 2

[1]	雷雨梦, 闫国超, 杨静, 朱祝军. 青花菜中矿质元素的近红外光谱快速测定[J]. 浙江农业学报, 2022, 34(3): 548-556.
[2]	张矞勋, 王磊, 璩向宁, 曹媛, 吴梦瑶, 于瑞鑫, 孙源. GF-1/WFV在玉米叶面积指数估算中的应用研究[J]. 浙江农业学报, 2021, 33(5): 861-872.
[3]	孙玉婷, 王映龙, 杨红云, 周琼, 孙爱珍, 杨文姬. RGB与HSI色彩空间下预测叶绿素相对含量的研究[J]. 浙江农业学报, 2018, 30(10): 1782-1788.
[4]	潘冰燕1，鲁晓翔1，*，张鹏2，李江阔2，陈绍慧2. 基于改进偏最小二乘法的甜椒可溶性固形物和总酸的近红外光谱检测[J]. 浙江农业学报, 2015, 27(10): 1739-.