Comparison of Machine Learning Algorithms for Remote Sensing Monitoring of Rice Yields

Xia JING; Jie ZHANG; Jiao-jiao WANG; Shi-kang MING; You-qiang FU; Hai-kuan FENG; Xiao-yu SONG

doi:10.3964/j.issn.1000-0593(2022)05-1620-08

Journals >Spectroscopy and Spectral Analysis >Volume 42 >Issue 5 >Page 1620 > Article

Spectroscopy and Spectral Analysis
Vol. 42, Issue 5, 1620 (2022)

Comparison of Machine Learning Algorithms for Remote Sensing Monitoring of Rice Yields

Xia JING^1、1;, Jie ZHANG^{1、1; 2;}, Jiao-jiao WANG^2、2;, Shi-kang MING^2、2;, You-qiang FU^3、3;, Hai-kuan FENG^2、2;, and Xiao-yu SONG^{2、2; *;}

Author Affiliations

¹1. School of Surveying and Mapping Science and Technology, Xi'an University of Science and Technology, Xi'an 710054, China

²2. Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100094, China

³3. Rice Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, China

show less

DOI: 10.3964/j.issn.1000-0593(2022)05-1620-08 Cite this Article

Xia JING, Jie ZHANG, Jiao-jiao WANG, Shi-kang MING, You-qiang FU, Hai-kuan FENG, Xiao-yu SONG. Comparison of Machine Learning Algorithms for Remote Sensing Monitoring of Rice Yields[J]. Spectroscopy and Spectral Analysis, 2022, 42(5): 1620 Copy Citation Text

show less

Fig. 1. Correlation between canopy spectra and growth parameters at different growth stages of rice in 2019 and 2020

Download full size | View in the Article

Fig. 2. Correlation between canopy spectra and growth parameters at different growth stages of rice in 2020

Download full size | View in the Article

Fig. 3. Yield prediction model based on full band spectra

Download full size | View in the Article

Fig. 4. Yield prediction model based on full band spectrum-growth parameters

Download full size | View in the Article

Fig. 5. Yield prediction model based on full band spectrum

Download full size | View in the Article

Fig. 6. Yield prediction model based on full band spectrum-growth parameters

Download full size | View in the Article

Fig. 7. Yield prediction model based on full band spectrum-growth parameters-crop nutrient absorption

Download full size | View in the Article

参数	2019年		2020年
参数	分化期	抽穗期	分化期	抽穗期
冠层光谱	√	√	√	√
长势参数(AGB, LAI)	√	√	√	√
土壤养分参数	未获取		√	√

Table 1. Test data acquisition

View in the Article

作物养分吸收量/ (kg·hm^-2)	相关系数r
	产量	AGB		LAI
	产量	分化期	抽穗期	分化期	抽穗期
至抽穗期作物吸收N量	0.713^**	0.797^**	0.623^**	0.789^**	0.731^**
至抽穗期作物吸收P量	-0.086	-0.085	-0.181	-0.093	-0.098
至抽穗期作物吸收K量	0.526^*	0.536^*	0.425	0.591^**	0.490^*
至成熟期作物吸收N量	0.723^**	0.595^**	0.609^**	0.818^**	0.716^**
至成熟期作物吸收P量	-0.242	-0.288	-0.415	-0.254	-0.378
至成熟期作物吸收K量	0.402	0.380	0.275	0.478^*	0.341

Table 2. Correlation between biomass, LAI, yield and crop nutrient uptake in 2020

View in the Article

参数	模型	精度
		分化期		抽穗期
		R²	NRMSE /%	R²	NRMSE /%
	BRR	0.897 1	6.76	0.864 7	7.76
全波段	SVR	0.877 6	7.59	0.865 5	7.82
	PLSR	0.754 1	10.44	0.555 1	14.04
	BRR	0.903 0	6.57	0.915 2	6.24
全波段+AGB, LAI	SVR	0.850 1	8.22	0.870 9	8.01
	PLSR	0.752 0	10.48	0.560 7	13.95

Table 3. Model accuracies based on the data in 2019 and 2020

View in the Article

参数	模型	精度
		分化期		抽穗期
		R²	NRMSE /%	R²	NRMSE /%
全波段	BRR	0.902 2	5.54	0.849 3	6.89
	SVR	0.836 2	7.30	0.827 6	7.41
	PLSR	0.644 8	10.53	0.549 5	11.85
全波段+ AGB, LAI	BRR	0.925 0	4.86	0.905 7	5.45
	SVR	0.837 0	7.26	0.845 7	7.36
	PLSR	0.654 6	10.38	0.659 8	11.58
全波段+ AGB、 LAI+ 作物养分吸收量	BRR	0.940 3	4.34	0.922 4	4.95
	SVR	0.845 7	7.03	0.889 0	6.39
	PLSR	0.659 8	10.30	0.575 3	11.51

Table 4. Model construction accuracy based on the data in 2020

Download Citation

Save the article for my favorites

Paper Information