Construction and Verification of a Mathematical Model for Near-Infrared Spectroscopy Analysis of Gel Consistency in Southern Indica Rice

Hong-mei LIU; Tao SHEN; Wen-yi ZHANG; Xi-wen SHI; Tao DAI; Tao BAI; Ying-hui XIAO

doi:10.3964/j.issn.1000-0593(2021)08-2432-05

Journals >Spectroscopy and Spectral Analysis >Volume 41 >Issue 8 >Page 2432 > Article

Spectroscopy and Spectral Analysis
Vol. 41, Issue 8, 2432 (2021)

Construction and Verification of a Mathematical Model for Near-Infrared Spectroscopy Analysis of Gel Consistency in Southern Indica Rice

Hong-mei LIU^*, Tao SHEN, Wen-yi ZHANG, Xi-wen SHI..., Tao DAI, Tao BAI and Ying-hui XIAO^*;|Show fewer author(s)

Author Affiliations

College of Agronomy, Hunan Agricultural University, Changsha 410128, China

show less

DOI: 10.3964/j.issn.1000-0593(2021)08-2432-05 Cite this Article

Hong-mei LIU, Tao SHEN, Wen-yi ZHANG, Xi-wen SHI, Tao DAI, Tao BAI, Ying-hui XIAO. Construction and Verification of a Mathematical Model for Near-Infrared Spectroscopy Analysis of Gel Consistency in Southern Indica Rice[J]. Spectroscopy and Spectral Analysis, 2021, 41(8): 2432 Copy Citation Text

show less

Fig. 1. Near infrared spectra of the calibration samples

Download full size | View in the Article

Fig. 2. The leverage and residual of rice gel consistency by near infrared analysis

Download full size | View in the Article

Fig. 3. The correlation of predicted value and chemical value of rice gel consistency established by regression model

Download full size | View in the Article

样本	样品数量	胶稠度/mm	平均值/mm	标准差/mm
校正集	107	43.0~94.0	71.0	12.2
验证集	30	43.5~88.0	70.8	11.5

Table 1. The chemical value of rice gel consistency

View in the Article

预处理	主因子	校正集			验证集			RPD
预处理	主因子	R	SEC	RMSEC	r	SECV	RMSECV	RPD
无处理	6	0.970 0	2.973 6	2.959 6	0.962 6	3.313 4	3.298 0	3.699 2
SNV	7	0.947 2	3.919 6	3.901 2	0.936 0	4.304 5	4.284 4	2.847 5
MSC	7	0.947 0	3.928 5	3.910 1	0.934 6	4.349 4	4.329 1	2.818 1
平滑	6	0.970 0	2.971 8	2.957 9	0.964 2	3.242 1	3.227 1	3.780 5
一阶导数	5	0.951 2	3.774 2	3.756 5	0.941 0	4.138 4	4.119 0	2.961 9
二阶导数	6	0.971 7	2.888 6	2.875 1	0.930 3	4.498 7	4.478 3	2.724 2
SNV+MSC	7	0.947 0	3.928 5	3.910 1	0.933 8	4.376 8	4.359 2	2.798 7
MSC+平滑	8	0.954 7	3.638 8	3.621 7	0.941 8	4.112 3	4.094 0	2.980 0
SNV+平滑	7	0.946 3	3.951 3	3.932 8	0.932 7	4.414 3	4.394 0	2.776 5
SNV+1阶	4	0.943 0	4.069 9	4.050 8	0.928 1	4.552 7	4.531 5	2.692 3
SNV+2阶	5	0.960 1	3.420 1	3.404 1	0.925 1	4.645 3	4.623 7	2.638 6
MSC+1阶	4	0.943 0	4.070 0	4.051 0	0.932 7	4.409 3	4.388 6	2.779 9
MSC+2阶	5	0.960 0	3.423 6	3.407 6	0.923 5	4.691 1	4.669 2	2.612 9
平滑+1阶	5	0.948 5	3.873 1	3.855 0	0.938 3	4.227 7	4.207 9	2.899 3
平滑+2阶	5	0.951 1	3.777 6	3.759 9	0.930 3	4.488 5	4.468 1	2.730 5
MSC+平滑+1阶	4	0.939 0	4.205 8	4.186 1	0.927 5	4.570 4	4.549 3	2.681 7
MSC+平滑+2阶	4	0.944 5	4.017 3	3.998 5	0.932 1	4.430 6	4.410 4	2.766 2
SNV+平滑+1阶	4	0.939 0	4.206 0	4.186 3	0.927 3	4.574 9	4.554 4	2.678 7
SNV+平滑+2阶	4	0.944 6	4.013 8	3.995 0	0.928 3	4.546 5	4.526 3	2.695 4
平滑+1阶+2阶	6	0.961 4	3.363 4	3.347 6	0.933 0	4.408 2	4.387 7	2.780 5

Table 2. Evaluation indexes of NIR model of rice gel consistency under different treatments

View in the Article

波长段/nm	主因子	校正集			验证集			RPD
波长段/nm	主因子	R	SEC	RMSEC	r	SECV	RMSECV	RPD
950~1 100	7	0.718 4	8.503 9	8.464 1	0.545 8	10.481 4	10.433 1	1.169 4
1 100~1 300	8	0.942 1	4.099 1	4.079 9	0.923 3	4.721 8	4.699 9	2.595 8
1 300~1 650	9	0.953 8	3.672 6	3.655 4	0.942 1	4.101 1	4.082 0	2.988 7
950~1 300	10	0.957 4	3.529 4	3.512 8	0.938 7	4.218 0	4.198 2	2.906 0
1 100~1 650	7	0.969 4	3.003 0	2.989 0	0.963 8	3.261 1	3.245 9	3.758 6
950~1 650	6	0.970 0	2.971 8	2.957 9	0.963 2	3.287 7	3.272 6	3.727 9

Table 3. Evaluation Indexes of near infrared analysis models of rice gel consistency at different wavelength bands

View in the Article

样品	预测值	化学值	绝对误差	样品	预测值	化学值	绝对误差
1	48.294 5	43.5	4.794 5	16	75.904 8	73.0	2.904 8
2	52.380 1	47.5	4.880 1	17	70.315 9	74.0	3.684 1
3	57.740 7	53.5	4.240 7	18	71.725 1	74.5	2.774 9
4	54.501 1	56.0	1.498 9	19	70.693 6	76.5	5.806 4
5	55.580 0	57.0	1.420 0	20	75.629 2	77.0	1.370 8
6	57.957 7	60.5	2.542 3	21	78.783 3	78.0	0.783 3
7	57.957 7	62.5	4.542 3	22	76.977 8	79.5	2.522 2
8	63.698 6	63.5	0.198 6	23	74.497 6	81.0	6.502 4
9	67.740 0	64.5	3.240 0	24	84.062 9	81.5	2.562 9
10	66.835 4	65.5	1.335 4	25	77.737 6	82.5	4.762 4
11	68.602 2	67.0	1.602 2	26	87.601 7	83.0	4.601 7
12	73.697 1	68.0	5.697 1	27	83.284 8	84.5	1.215 2
13	64.304 1	68.5	4.195 9	28	83.049 9	85.0	1.950 1
14	71.836 8	70.5	1.336 8	29	82.190 9	86.0	3.809 1
15	73.229 0	72.0	1.229 0	30	89.367 6	88.0	1.367 6

Table 4. Predicted value and chemical value of validation set

View in the Article

验证集样本(30个)
指标	t检验	自由度	sig.(双侧)	均值差值	95%置信区间
指标	t检验	自由度	sig.(双侧)	均值差值	下限	上限
统计值	-0.353	29	0.726	0.643 33	-1.543 17	1.088 35

Table 5. Paired t test of model prediction value and chemical value of rice gel consistency in validation set

Download Citation

Save the article for my favorites

Paper Information

微信扫一扫：分享

微信扫一扫：分享