基于高光谱的琯溪蜜柚叶片磷素含量估算模型研究

doi:10.13304/j.nykjdb.2021.1002

摘要/Abstract

摘要：

蜜柚叶片磷素（phosphorus，P）含量是准确诊断和定量评价生长状况的重要指标，为快速、无损、精确地估测磷素含量，需要建立蜜柚叶片磷素含量高光谱估算模型。基于蜜柚叶片高光谱数据和磷素含量实测数据，提取原始光谱及一阶微分光谱特征波段和光谱特征变量，构建单变量估算模型、偏最小二乘回归模型和BP神经网络回归模型，并确定蜜柚叶片磷素含量最佳估算模型。在350~1 050 nm波段，原始光谱和一阶微分光谱与叶片磷素含量在可见光范围内有多波段相关性显著，并出现多个极值。原始光谱敏感波长为549和718 nm，一阶微分的敏感波长为528、703和591 nm。在建立的回归模型中，选择决定系数较高的模型进行精度检验，其中BP神经网络模型的拟合R²（0.775 9）最大，偏最小二乘估算模型的拟合R²（0.749 9）次之。综合建模精度和模型检验精度，确定BP神经网络模型为蜜柚叶片磷含量的最佳估算模型，建模和验证的R²分别为0.71和0.775 9；其次为偏最小二乘估算模型，建模和验证的R²分别为0.64和0.749 9。上述结果可为大面积的蜜柚叶片营养遥感监测诊断和合理施肥提供理论依据。

关键词: 高光谱, 蜜柚, 磷素, 光谱指数

Abstract:

Phosphorus （P） content in honey pomelo leaves is an important indice for accurate diagnosis and quantitative evaluation of growth status. The hyperspectral estimation models of P contents in honey pomelo leaves were established to provide basis for rapid， non-destructive and accurate estimation of P content. Based on the hyperspectral data and the measured data of P contents in pomelo leaves， the characteristic bands and spectral characteristic variables of the original spectrum and the first-order differential spectrum were extracted， and the single variable estimation model， partial least squares regression model and BP neural network regression model were constructed， and the best estimation model of P content was determined. The reflectance spectra of pomelo leaves with different P contents were significantly different at 350~1 050 nm. The correlation between the original spectrum and the first-order differential spectrum and the leaves P contents in the visible light range were significant， and there were multiple extremums. The sensitive wavelengths of the original spectral curve were 549 and 718 nm， and the sensitive wavelengths of the first order differential curve were 528， 703 and 591 nm， respectively. Among the established regression models， the model with higher coefficient of determination was selected for precision test. The fitting R² （0.775 9） of BP neural network model was the largest， followed by the fitting R²（0.749 9） of partial least squares estimation model. The results showed that the BP neural network model was the best one to estimate the P contents of pomelo leaves， the determination coefficient （R²） of modeling and verification were 0.71 and 0.775 9， respectively； the second was the partial least squares estimation model， the R² of modeling and verification were 0.64 and 0.749 9， respectively. Above results provided a theoretical basis for the nutrition monitoring and diagnosis of large-area honey pomelo leaves and the rational fertilization.

Key words: hyperspectrum, honey pomelo, phosphorus, spectral index

中图分类号:

S666.3

栗方亮, 孔庆波, 张青. 基于高光谱的琯溪蜜柚叶片磷素含量估算模型研究[J]. 中国农业科技导报, 2023, 25(1): 100-108.

Fangliang LI, Qingbo KONG, Qing ZHANG. Estimation Models of Phosphorus Contents in Guanxi Honey Pomelo Leaves Based on Hyperspectral Data[J]. Journal of Agricultural Science and Technology, 2023, 25(1): 100-108.

图/表 5

参考文献 28

1	平和地方志研究室.平和年鉴(2020)[C].福州:海峡书局, 2021:167.
2	叶林蔚,唐荣年,李创.基于AE-FFNN神经网络的橡胶树叶片磷含量定性研究[J].中国农业科技导报,2021,23(7):117-124.
	YE L W, TANG R N, LI C. Qualitative study on phosphorus content in rubber leaves based on AE-FFNN neural work [J]. J. Agric. Sci. Technol., 2021,23 (7):117-124.
3	SIEDLISKA A, BARANOWSKI P, PASTUSZKA-WOŹNIAK J, et al.. Identification of plant leaf phosphorus content at different growth stages based on hyperspectral reflectance [J]. BMC Plant Biol., 2021, 21(1):28-45.
4	李岚涛,汪善勤,任涛,等.基于高光谱的冬油菜叶片磷含量诊断模型[J].农业工程学报,2016,32(14):209-218.
	LI L T, WANG S Q, REN T, et al.. Evaluating models of leaf phosphorus content of winter oilseed rape based on hyperspectral data [J]. Trans. Chin. Soc. Agric. Eng., 2016, 32(14): 209-218.
5	VIÉGAS I D J M, CORDEIRO R A M, ALMEIDA G M D, et al.. Growth and visual symptoms of nutrients deficiency in Mangos teens (Garcinia mangostana L.) [J]. Am. J. Plant Sci., 2018,5:1014-1028.
6	DUNN B L, SINGH H, PAYTON M, et al.. Effects of nitrogen, phosphorus, and potassium on SPAD-502 and at LEAF sensor readings of Salvia [J]. J. Plant Nutr., 2018,41(13):1674-1683.
7	栗方亮,孔庆波,张青.光谱分析技术在作物氮素诊断上的应用研究进展[J].江西农业学报,2020,32(9):6-12.
	LI F L, KONG Q B, ZHANG Q. Research progress in application of spectral analysis technology in crop nitrogen diagnosis [J]. Acta Agric. Jiangxi, 2020, 32(9):6-12.
8	KOVAR M, BRESTIC M, SYTAR O, et al.. Evaluation of hyperspectral reflectance parameters to assess the leaf water content in soybean [J]. Water, 2019, 443(11):443-456.
9	LI F, WANG L, LIU J, et al.. Evaluation of leaf N concentration in winter wheat based on discrete wavelet transform analysis [J]. Remote Sens., 2019, 11(11): 1331-1341.
10	SYTAR O, ZIVCAK M, NEUGART S, et al.. Assessment of hyperspectral indicators related to the content of phenolic compounds and multispectral fluorescence records in chicory leaves exposed to various light environments [J]. Plant Physiol. Biochem., 2020,154: 429-438.
11	WANG J, WANG T, SKIDMORE A K, et al.. Evaluating different methods for grass nutrient estimation from canopy hyperspectral reflectance [J]. Remote Sens., 2015,7(5): 5901-5917.
12	MAHAJAN G R, SAHOO R N, PANDEY R N, et al.. Using hyperspectral remote sensing techniques to monitor nitrogen, phosphorus, sulphur and potassium in wheat (Triticum aestivum L.) [J]. Precision Agric., 2014,15: 499-522.
13	冯海宽,杨福芹,李振海,等.最优权重组合模型和高光谱估算苹果叶片全磷含量[J].农业工程学报,2016,32(7):173-180.
	FENG H K, YANG F Q, LI Z H, et al.. Hyperspectral estimation of leaf total phosphorus content in apple tree based on optimal weights combination model [J]. Trans. Chin. Soc. Agric. Eng., 2016, 32(7): 173-180.
14	程立真,朱西存,高璐璐,等.基于随机森林模型的苹果叶片磷素含量高光谱估测[J].果树学报,2016,33(10):1219-1229.
	CHENG L Z, ZHU X C, GAO L L, et al.. Hyperspectral estimation of phosphorus content for apple leaves based on the random forest model [J]. J. Fruit Sci., 2016,33(10):1219-1229.
15	LIU J, HAN J C, CHEN X, et al.. Nondestructive detection of rape leaf chlorophyll level based on Vis-NIR spectroscopy [J/OL]. Spectrochim. Acta A Mol. Biomol. Spectrosc., 2019, 222: ‍117202 ‍[2022-‍10-‍22]. ‍2.
16	鲁如坤.土壤农业化学分析方法[M].北京:中国农业科学技术出版社,2000:312-314.
	LU R K. Agrochemical Analysis Methods of Soil [M]. Beijing: China Agricultural Science and Technology Press, 2000: 312-314.
17	刘冰峰,李军,贺佳.玉米叶片全磷含量高光谱遥感监测诊断模型研究[J].农业机械学报,2015,46(8):252-280.
	LIU B F, LI J, HE J. Total phosphorus content estimation models of summer maize leaves based on hyperspectral remote sensing [J]. Trans. Chin. Soc. Agric. Mach., 2015, 46(8): 252-280.
18	潘蓓,赵庚星,朱西存,等.基于高光谱的苹果树冠层磷素状况估测模型研究[J].红外,2012,33(6):27-31.
	PAN B, ZHAO G X, ZHU X C, et al.. Estimation of phosphorus content in apple tree canopy based on hyperspectral [J]. Infrared, 2012, 33(6): 27-31.
19	岳学军,全东平,洪添胜,等.不同生长期柑橘叶片磷含量的高光谱预测模型[J].农业工程学报,2015,31(8):207-213.
	YUE X J, QUAN D P, HONG T S, et al.. Prediction model of phosphorus content for citrus leaves during different growth periods based on hyperspectrum [J]. Trans. Chin. Soc. Agric. Eng., 2015, 31(8): 207-213.
20	柴仲平,王雪梅,陈波浪,等.库尔勒香梨叶片全磷含量高光谱估算模型研究[J].西部林业科学,2013,42(5):7-12.
	CHAI Z P, WANG X M, CHEN B L, et al.. Hyperspectral estimation models for total phosphorus content of Kuerle Fragrant pear leaves [J]. J. West China For. Sci., 2013, 42(5): 7-12.
21	WANG J, SHI T, LIU H, et al.. Successive projections algorithm-based three-band vegetation index for foliar phosphorus estimation [J]. Ecol. Indic., 2016, 67:12-20.
22	邢东兴,常庆瑞.基于光谱分析的果树叶片全氮、全磷、全钾含量估测研究-以红富士苹果树为例[J].西北农林科技大学学报(自然科学版),2009,37(2):141-154.
	XING D X, CHANG Q R. Research on predicting the TN, TP, TK contents of fresh fruit tree leaves by spectral analysis with Red Fuji Apple tree as an example [J]. J. Northwest A&F Univ. (Nat. Sci.), 2009, 37(2): 141-154.
23	易时来,邓烈,何绍兰,等.基于冠层光谱的锦橙叶片磷素营养监测研究[J].核农学报,2013,27(2):225-230.
	YI S L, DENG L, HE S L, et al.. Monitoring leaf phosphorus nutrition based on canopy spectrum of Jincheng orange[Citrus sinensis ( L. )] [J]. Acta Agric. Nucl. Sin., 2013, 27(2):225-230.
24	朱西存,赵庚星,董芳,等.基于高光谱的苹果花磷素含量监测模型[J].应用生态学报,2009,20(10):2424-2430.
	ZHU X C, ZHAO G X, DONG F, et al.. Monitoring models for phosphorus content of apple flowers based on hyperspectrum [J]. Chin. J. Appl. Ecol., 2009, 20(10): 2424-2430.
25	刘云根,余哲修,张超,等.基于高光谱技术剑湖湿地茭草磷含量估算模型研究[J].西南林业大学学报(自然科学),2019,39(1): 123-131.
	LIU Y G, YU Z X, ZHANG C, et al..The estimation model for phosphorus content of Zizania Cuciflora in Jianhu wetland based on hyperspectral technology [J]. J. Southwest For. Univ., 2019, 39(1): 123-131.
26	黄双萍,洪添胜,岳学军,等.基于高光谱的柑橘叶片磷含量估算模型实验[J].农业机械学报,2013,44(4):202-207.
	HUANG S P, HONG T S, YUE X J,et al.. Hyperspectral estimation model of total phosphorus content for citrus leaves [J]. Trans. Chin. Soc. Agric. Mach., 2013,44(4):202-207.
27	CHEN L, HUANG J F, WANG F M, et al.. Comparison between back propagation neural network and regression models for the estimation of pigment content in rice leaves and panicles using hyperspectral data [J]. Int. J. Remote Sens., 2007, 28(16):3457-3478.
28	高洪燕,毛罕平,张晓东.生菜叶中磷含量的光谱定量分析[J].农业机械学报,2014,45(增):276-280.
	GAO H Y, MAO H P, ZHANG X D. Quantitative determination of phosphorus in lettuce leaf using spectroscopy [J]. Trans. Chin. Soc. Agric. Mach., 2014, 45(S):276-280.

	回归模型方式 Regression model method	R²	RMSE	RE/%
单变量回归 Univariate regression	NDVI′_591，703	0.649 1	0.17	9.50
	RVI′_591，703	0.648 7	0.17	9.48
	NDVI_549，718	0.602 3	1.23	8.56
	RVI_549，718	0.606 7	0.15	8.51
	偏最小二乘法PLS	0.749 9	0.14	7.11
	BP神经网络BPNN	0.775 9	0.13	6.99

	回归模型方式 Regression model method	R²	RMSE	RE/%
单变量回归 Univariate regression	NDVI′_591，703	0.649 1	0.17	9.50
	RVI′_591，703	0.648 7	0.17	9.48
	NDVI_549，718	0.602 3	1.23	8.56
	RVI_549，718	0.606 7	0.15	8.51
	偏最小二乘法PLS	0.749 9	0.14	7.11
	BP神经网络BPNN	0.775 9	0.13	6.99

[1]	王靖会, 程娇娇, 刘洋, 常佳乐, 王朝辉 . 基于高光谱成像技术鉴别大米品种[J]. 中国农业科技导报, 2021, 23(9): 121-128.
[2]	于洋, 张珏, 田海清, 王迪, 王轲, 张红旗. 基于高光谱成像技术的冷鲜羊肉嫩度检测方法研究[J]. 中国农业科技导报, 2021, 23(12): 101-108.
[3]	赵宁博, 杨佳佳, 赵英俊, 秦凯, 杨越超, 李明. 基于航空高光谱反演的黑土地质量综合评价研究[J]. 中国农业科技导报, 2021, 23(11): 88-98.
[4]	王靖会1,曹崴1,冷全阳1,程娇娇1,王艳辉2,沈海鸥3,陈雷4,王朝辉5*. 基于高光谱与在线序列极限学习机确证大米产地方法[J]. 中国农业科技导报, 2020, 22(9): 96-103.
[5]	郭晗1,2,张序1*,陆洲2,田婷3,徐飞飞2,罗明2,吴正贵4,孙振军5. 基于机载高光谱影像的南方水稻土有机质含量估算[J]. 中国农业科技导报, 2020, 22(6): 60-71.
[6]	杨婧，廖桂平*，刘凡，官春云. 基于高光谱成像技术的油菜叶片叶绿素含量预测[J]. 中国农业科技导报, 2020, 22(5): 86-96.
[7]	杨菲菲，李世娟*，刘升平，吕纯阳，刘大众，肖顺夫，刘航. 作物环境胁迫高光谱遥感监测研究进展[J]. 中国农业科技导报, 2020, 22(4): 85-93.
[8]	沈宇1,2，房胜2，王风云1，李哲2，张琛1,2，郑纪业1*. 基于高光谱成像技术识别苹果轻微损伤的有效波段研究[J]. 中国农业科技导报, 2020, 22(3): 64-71.
[9]	任永新,单忠德,张静,战丽. 计算机视觉技术在水果品质检测中的研究进展[J]. , 2012, 14(1): 98-103.
[10]	孙传范1,肖凯2,韩胜芳2,赵芳华2,鲍金香2. 植物吸收和转运磷素的分子机理研究进展[J]. , 2011, 13(2): 17-24.
[11]	李刚,杨志勇,钦佩. 高光谱遥感数据估算蓖麻初级生产力主要参数研究[J]. , 2009, 11(5): 114-118.