Cotton Science

Figure/Table detail

Development of a rapid non-destructive method for the detection of cottonseed oil content by near-infrared spectroscopy

Tian Boyu, Huang Yiwen, Zhou Dayun, Huang Longyu, Wu Yuzhen, Fu Shouyang, Xu Qing, Guo Jinggong, Peng Jun, Kuang Meng

Cotton Science, 2023, 35(4): 325-333. DOI: 10.11963/cs20230005

模型编号 Model number	数学处理 Mathematical processing	棉籽仁Kernel				光籽 Delinted seed				毛籽 Linted seed
模型编号 Model number	数学处理 Mathematical processing	SEC	RSQ	SECV	1-VR	SEC	RSQ	SECV	1-VR	SEC	RSQ	SECV	1-VR
1	1, 4, 4, 1	0.83	0.96	1.12	0.92	1.10	0.93	1.33	0.89	1.39	0.90	1.94	0.80
2	2, 4, 4, 1	1.04	0.93	1.10	0.92	0.93	0.95	1.16	0.92	1.40	0.89	1.85	0.82
3	3, 4, 4, 1	0.87	0.95	1.03	0.94	1.09	0.93	1.27	0.91	1.64	0.84	1.83	0.81
4	4, 4, 4, 1	1.05	0.93	1.19	0.91	1.15	0.92	1.42	0.88	1.36	0.90	1.83	0.81
5	1, 5, 5, 1	0.83	0.96	1.14	0.92	1.09	0.94	1.47	0.89	1.31	0.92	1.98	0.80
6	2, 5, 5, 1	0.72	0.97	1.01	0.94	0.99	0.95	1.20	0.92	1.35	0.90	1.88	0.83
7	3, 5, 5, 1	0.97	0.94	1.04	0.93	1.07	0.93	1.25	0.91	1.61	0.86	1.99	0.78
8	4, 5, 5, 1	0.93	0.95	1.10	0.93	1.13	0.93	1.33	0.90	1.20	0.91	1.72	0.82
9	1, 6, 6, 1	1.11	0.93	1.25	0.91	1.04	0.94	1.31	0.91	1.44	0.90	1.87	0.83
10	2, 6, 6, 1	0.86	0.95	1.06	0.92	1.13	0.93	1.27	0.91	1.23	0.92	1.82	0.82
11	3, 6, 6, 1	1.05	0.93	1.13	0.92	1.14	0.93	1.30	0.91	1.33	0.90	1.79	0.83
12	4, 6, 6, 1	0.82	0.96	0.97	0.94	1.15	0.92	1.36	0.89	1.39	0.89	1.85	0.80
13	1, 7, 7, 1	0.95	0.94	1.07	0.93	1.16	0.93	1.49	0.88	1.62	0.85	2.05	0.77
14	2, 7, 7, 1	0.95	0.95	0.98	0.95	1.10	0.93	1.24	0.92	1.19	0.92	1.68	0.85
15	3, 7, 7, 1	0.85	0.95	1.01	0.93	0.94	0.95	1.13	0.93	1.17	0.93	1.68	0.84
16	4, 7, 7, 1	0.81	0.96	1.12	0.92	1.07	0.94	1.27	0.91	1.25	0.93	1.69	0.84
17	1, 8, 8, 1	0.93	0.94	1.10	0.92	1.14	0.93	1.48	0.89	1.90	0.81	2.34	0.71
18	2, 8, 8, 1	1.05	0.93	1.08	0.93	1.16	0.93	1.38	0.90	1.33	0.91	1.82	0.82
19	3, 8, 8, 1	0.83	0.96	1.05	0.93	0.94	0.95	1.26	0.92	1.32	0.91	1.69	0.85
20	4, 8, 8, 1	0.71	0.97	1.00	0.94	1.08	0.93	1.25	0.91	1.21	0.93	2.07	0.79
21	1, 9, 9, 1	0.90	0.94	1.02	0.93	1.10	0.94	1.52	0.88	1.90	0.82	2.38	0.72
22	2, 9, 9, 1	1.23	0.92	1.19	0.92	1.08	0.94	1.43	0.89	1.23	0.93	1.81	0.84
23	3, 9, 9, 1	0.71	0.97	0.95	0.95	1.03	0.94	1.22	0.92	1.08	0.94	1.62	0.86
24	4, 9, 9, 1	0.73	0.97	0.96	0.94	1.20	0.92	1.45	0.88	1.14	0.93	1.69	0.84

Table 1 Evaluation parameters of oil content models for different mathematical processing of three morphological types of cottonseed

Other figure/table from this article

Fig. 1 Distribution of cottonseed oil content in calibration sample and verification sample A: Distribution of cottonseed oil content in calibration samples; B: Distribution of cottonseed oil content in verification samples.
Fig. 2 NIR spectra of cottonseed kernel, naked seed and gross seed A, C, E: Original NIR spectra of cotton seed kernel, delinted seed and linted seed; B, D, F: Processed spectra of cotton seed kernel, delinted seed and linted seed, and the data processing method were the SNV plus DET combined with mathematical parameters of 4, 6, 6, 1; 3, 9, 9, 1; 4, 7, 7, 1.
Table 2 Verification results for oil content predictation model by validation samples
Fig. 3 Correlation of oil content value between NIRS model predition and chemical method detected in validation sample set A: Optimal calibration model of oil content in kernel; B: Optimal calibration model of oil content in delinted seed; C: Optimal calibration model of oil content in linted seed.
Fig. 4 Distribution of oil content difference between actual value and predicted value in validation sample set

模态框（Modal）标题

Figure/Table detail