Hyperspectral Inversion Model of Pectin Content in Wheat Under Salt and Physical Damage Stresses

Zhao-jia PIAO; Hai-ye YU; Jun-he ZHANG; Hai-gen ZHOU; Shuang LIU; Li-juan KONG; Jing-min DANG

doi:10.3964/j.issn.1000-0593(2022)09-2935-06

Journals >Spectroscopy and Spectral Analysis >Volume 42 >Issue 9 >Page 2935 > Article

Spectroscopy and Spectral Analysis
Vol. 42, Issue 9, 2935 (2022)

Hyperspectral Inversion Model of Pectin Content in Wheat Under Salt and Physical Damage Stresses

Zhao-jia PIAO^*, Hai-ye YU, Jun-he ZHANG, Hai-gen ZHOU, Shuang LIU, Li-juan KONG, and Jing-min DANG^*;

Author Affiliations

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

show less

DOI: 10.3964/j.issn.1000-0593(2022)09-2935-06 Cite this Article

Zhao-jia PIAO, Hai-ye YU, Jun-he ZHANG, Hai-gen ZHOU, Shuang LIU, Li-juan KONG, Jing-min DANG. Hyperspectral Inversion Model of Pectin Content in Wheat Under Salt and Physical Damage Stresses[J]. Spectroscopy and Spectral Analysis, 2022, 42(9): 2935 Copy Citation Text

show less

Changes of pectin content in wheat leaves under different stress factors in the first week(a): Under single stress factor; (b): Under multiple stress factors

Fig. 1. Changes of pectin content in wheat leaves under different stress factors in the first week
(a): Under single stress factor; (b): Under multiple stress factors

Download full size | View in the Article

Average pectin contents in wheat leaves corresponding to different stress factors in the first weekDifferent lowercase letters indicate the significant difference between different treatments at p<0.05 (a): Single stress factor; (b): Multiple stress factors

Fig. 2. Average pectin contents in wheat leaves corresponding to different stress factors in the first week
Different lowercase letters indicate the significant difference between different treatments at p<0.05 (a): Single stress factor; (b): Multiple stress factors

Download full size | View in the Article

Fig. 3. Hyperspectral characteristics of wheat leaves under different stress factors
(a): Under single stress factor; (b): Under multiple stress factors

Download full size | View in the Article

Fig. 4. Characteristic wavelength extraction by correlation coefficient method
(a): Correlation between original spectra and pectin content in wheat leaves; (b): Correlation between first-derivative spectra and pectin content in wheat leaves

Download full size | View in the Article

Fig. 5. (a) The relationship between the pectin content predicted by the optimal model and the pectin content measured by the kit; (b) Precision evaluation of the proposed pectin prediction model

Download full size | View in the Article

	分组	盐胁迫	物理损伤
单因素胁迫实验	A1	50 mmol·L^-1 NaCl	-
	A2	100 mmol·L^-1 NaCl	-
	A3	150 mmol·L^-1 NaCl	-
	S1	-	针刺10孔
	S2	-	针刺30孔
	CK	-	-
复合胁迫实验	A1S1	50 mmol·L^-1 NaCl	针刺10孔
	A1S2	50 mmol·L^-1 NaCl	针刺30孔
	A2S1	100 mmol·L^-1 NaCl	针刺10孔
	A2S2	100 mmol·L^-1 NaCl	针刺30孔
	A3S1	150 mmol·L^-1 NaCl	针刺10孔
	A3S2	150 mmol·L^-1 NaCl	针刺30孔
	CK	-	-

Table 1. Experimental design of salt and physical damage stresses

View in the Article

建模对象	建模方法	预处理方法	R_c	R_p
果胶含量	PCR	MSC+SG(W_702~728)	0.902 4	0.880 1
		SNV+SG(W_702~728)	0.917 6	0.908 7
		MSC+FD+SG(W_690~700)	0.906 6	0.877 8
		MSC+FD+NDF(W_690~700)	0.941 1	0.950 6
		SNV+FD+SG(W_690~700)	0.927 4	0.920 4
		SNV+FD+NDF(W_690~700)	0.936 0	0.943 8
		FD(W_690~700)	0.851 5	0.860 6
		W_702~728	0.872 5	0.854 8
		无	0.862 4	0.821 1
果胶含量	PLS	MSC+SG(W_702~728)	0.914 5	0.918 8
		SNV+SG(W_702~728)	0.912 1	0.906 4
		MSC+FD+SG(W_690~700)	0.961 2	0.938 2
		MSC+FD+NDF(W_690~700)	0.948 8	0.896 7
		SNV+FD+SG(W_690~700)	0.952 0	0.930 4
		SNV+FD+NDF(W_690~700)	0.978 2	0.981 4
		FD(W_690~700)	0.917 4	0.905 8
		W_702~728	0.887 6	0.893 4
		无	0.876 8	0.889 2
果胶含量	SMLR	MSC+SG(W_702~728)	0.876 5	0.877 8
		SNV+SG(W_702~728)	0.874 8	0.890 2
		MSC+FD+SG(W_690~700)	0.900 3	0.899 6
		MSC+FD+NDF(W_690~700)	0.871 4	0.902 8
		SNV+FD+SG(W_690~700)	0.802 8	0.788 6
		SNV+FD+NDF(W_690~700)	0.791 0	0.768 4
		FD(W_690~700)	0.785 5	0.817 3
		W_702~728	0.814 0	0.828 0
		无	0.779 1	0.773 6

Table 2. Prediction models obtained by different modeling methods and preprocessing techniques

Download Citation

Save the article for my favorites

Paper Information