Hyperspectral Analysis of Milk Protein Content Using SVM Optimized by Sparrow Search Algorithm

Mei-chen LIU; He-ru XUE; Jiang-ping LIU; Rong-rong DAI; Peng-wei HU; Qing HUANG; Xin-hua JIANG

doi:10.3964/j.issn.1000-0593(2022)05-1601-06

Journals >Spectroscopy and Spectral Analysis >Volume 42 >Issue 5 >Page 1601 > Article

Spectroscopy and Spectral Analysis
Vol. 42, Issue 5, 1601 (2022)

Hyperspectral Analysis of Milk Protein Content Using SVM Optimized by Sparrow Search Algorithm

Mei-chen LIU^*, He-ru XUE^*;, Jiang-ping LIU, Rong-rong DAI, Peng-wei HU, Qing HUANG, and Xin-hua JIANG

Author Affiliations

College of Computer and Information Engineering, Inner Mongolia Agricultural University, Huhhot 010000, China

show less

DOI: 10.3964/j.issn.1000-0593(2022)05-1601-06 Cite this Article

Mei-chen LIU, He-ru XUE, Jiang-ping LIU, Rong-rong DAI, Peng-wei HU, Qing HUANG, Xin-hua JIANG. Hyperspectral Analysis of Milk Protein Content Using SVM Optimized by Sparrow Search Algorithm[J]. Spectroscopy and Spectral Analysis, 2022, 42(5): 1601 Copy Citation Text

show less

Fig. 1. High spectrometer

Download full size | View in the Article

Fig. 2. The original and pretreated spectra

Download full size | View in the Article

Fig. 3. Flow chart of SVM parameters optimized by SSA

Download full size | View in the Article

Fig. 4. RMSE vs number of characteristic wavelengths

Download full size | View in the Article

Fig. 5. Characteristic variable selection by CARS

Download full size | View in the Article

Fig. 6. Fitness curve

Download full size | View in the Article

Fig. 7. CARS-SSA-SVM verification results

Download full size | View in the Article

建模方法	RMSEP	$R_{p}^{2}$	RMSEC	$R_{c}^{2}$
N-PLS	0.313 6	0.924 4	0.129 3	0.963 5
Standardization-PLS	0.286 9	0.927 7	0.120 7	0.965 4
MSC-PLS	0.261 6	0.931 6	0.105 4	0.968 9

Table 1. Precision comparison of preprocessing models

View in the Article

建模方法	变量数	RMSE	R²	Time/s
Full-PLS	125	0.091 4	0.930 6	6.132 1
CARS-PLS	27	0.037 6	0.959 8	5.755 4
SPA-PLS	17	0.052 7	0.956 1	5.269 7
Full-SVM	125	0.105 4	0.928 6	4.231 1
CARS-SVM	27	0.040 1	0.957 1	3.987 1
SPA-SVM	17	0.053 3	0.948 6	3.236 7
Full-GA-SVM	125	0.029 7	0.962 1	16.326 9
CARS-GA-SVM	27	0.015 3	0.979 8	13.143 2
SPA-GA-SVM	17	0.023 1	0.968 7	10.052 7
Full-PSO-SVM	125	0.088 4	0.936 9	13.241 1
CARS-PSO-SVM	27	0.031 6	0.964 9	10.231 1
SPA-PSO-SVM	17	0.037 6	0.959 8	8.534 1
Full-SSA-SVM	125	0.009 2	0.988 9	6.056 9
CARS-SSA-SVM	27	0.001 1	0.999 6	4.112 1
SPA-SSA-SVM	17	0.004 8	0.998 4	4.052 7

Table 2. Precision comparison of different models

Download Citation

Save the article for my favorites

Paper Information