Study on Extracting Characteristic Wavelength of Soybean Physiological Information Based on Hyperspectral Technique

Shuang LIU; Hai-ye YU; Zhao-jia PIAO; Mei-chen CHEN; Tong YU; Li-juan KONG; Lei ZHANG; Jing-min DANG; Yuan-yuan SUI

doi:10.3964/j.issn.1000-0593(2020)11-3542-07

Journals >Spectroscopy and Spectral Analysis >Volume 40 >Issue 11 >Page 3542 > Article

Spectroscopy and Spectral Analysis
Vol. 40, Issue 11, 3542 (2020)

Study on Extracting Characteristic Wavelength of Soybean Physiological Information Based on Hyperspectral Technique

Shuang LIU, Hai-ye YU, Zhao-jia PIAO, Mei-chen CHEN, Tong YU, Li-juan KONG, Lei ZHANG, Jing-min DANG, and Yuan-yuan SUI

Author Affiliations

School of Biological and Agricultural Engineering, Jilin University, Changchun 130022, China

show less

DOI: 10.3964/j.issn.1000-0593(2020)11-3542-07 Cite this Article

Shuang LIU, Hai-ye YU, Zhao-jia PIAO, Mei-chen CHEN, Tong YU, Li-juan KONG, Lei ZHANG, Jing-min DANG, Yuan-yuan SUI. Study on Extracting Characteristic Wavelength of Soybean Physiological Information Based on Hyperspectral Technique[J]. Spectroscopy and Spectral Analysis, 2020, 40(11): 3542 Copy Citation Text

show less

Characteristic wavelengths selected by CARS algorithm(a): Chlorophyll content (D1); (b): Chlorophyll content(D2);(c): Light energy utilization (D1); (d): Light energy utilization(D2)

Fig. 1. Characteristic wavelengths selected by CARS algorithm
(a): Chlorophyll content (D1); (b): Chlorophyll content(D2);(c): Light energy utilization (D1); (d): Light energy utilization(D2)

Download full size | View in the Article

Characteristic wavelengths selected by SPA algorithm(a): Chlorophyll content (D1); (b): Chlorophyll content(D2);(c): Light energy utilization (D1); (d): Light energy utilization(D2)

Fig. 2. Characteristic wavelengths selected by SPA algorithm
(a): Chlorophyll content (D1); (b): Chlorophyll content(D2);(c): Light energy utilization (D1); (d): Light energy utilization(D2)

Download full size | View in the Article

Fig. 3. Characteristic wavelengths selected by CC algorithm
(a): Chlorophyll content (D1); (b): Chlorophyll content(D2);(c): Light energy utilization (D1); (d): Light energy utilization(D2)

Download full size | View in the Article

建模对象	日期	预处理方法	R_c	R_p	建模对象	日期	预处理方法	R_c	R_p
叶绿素含量	D1	RAW	0.842	0.828	光能利用率	D1	RAW	0.862	0.825
		MSC	0.869	0.846			MSC	0.887	0.850
		SNV	0.867	0.839			SNV	0.891	0.854
		SG	0.855	0.838			SG	0.894	0.821
		FD	0.858	0.824			FD	0.885	0.856
		SD	0.859	0.818			SD	0.862	0.827
		MSC-SG-FD	0.909	0.882			MSC-SG-FD	0.900	0.878
		MSC-SG-SD	0.893	0.833			MSC-SG-SD	0.891	0.887
		SNV-SG-FD	0.896	0.870			SNV-SG-FD	0.913	0.894
		SNV-SG-SD	0.903	0.838			SNV-SG-SD	0.907	0.862
	D2	RAW	0.875	0.730		D2	RAW	0.869	0.838
		MSC	0.903	0.876	MSC			0.881	0.856
		SNV	0.907	0.867	SNV			0.882	0.857
		SG	0.887	0.869	SG			0.889	0.787
		FD	0.883	0.814	FD			0.886	0.833
		SD	0.873	0.788	SD			0.870	0.823
		MSC-SG-FD	0.909	0.880	MSC-SG-FD			0.896	0.808
		MSC-SG-SD	0.899	0.873	MSC-SG-SD			0.900	0.865
		SNV-SG-FD	0.909	0.816	SNV-SG-FD			0.902	0.869
		SNV-SG-SD	0.897	0.869	SNV-SG-SD			0.900	0.866

Table 1. Optimization results from hyperspectral pre-processing methods

View in the Article

建模对象	日期	建模方法	变量保留数	P_cs	R_c	R_p	建模对象	日期	建模方法	变量保留数	P_cs	R_c	R_p
叶绿素含量	D1	PLS	512	10	0.909	0.882	光能利用率	D1	PLS	512	10	0.913	0.894
		CC-PLS	221	9	0.911	0.906			CC-PLS	234	9	0.919	0.902
		CARS-PLS	41	9	0.927	0.892			CARS-PLS	46	9	0.921	0.909
		SPA-PLS	20	7	0.944	0.911			SPA-PLS	27	7	0.929	0.912
	D2	PLS	512	10	0.909	0.880		D2	PLS	512	10	0.902	0.869
		CC-PLS	97	9	0.902	0.898			CC-PLS		224	9	0.907	0.885
		CARS-PLS	96	9	0.914	0.899	CARS-PLS			32	9	0.912	0.898
		SPA-PLS	23	7	0.941	0.903	SPA-PLS			37	7	0.925	0.907

Table 2. Soybean physiological information inversion model results

Download Citation

Save the article for my favorites

Paper Information