NIR Model Optimization Study of Larch Wood Density Based on IFSR Abnormal Sample Elimination

Table 1. Statistical analysis of correction set and prediction set results (g·cm^-3)
View table
View in Article
Table 1. Statistical analysis of correction set and prediction set results (g·cm^-3)
样本集样本数/个均值最大值最小值标准差
校正集 121 0.521 3 0.743 5 0.411 0 0.066 2
预测集 60 0.525 8 0.620 6 0.413 0 0.066 3

Table 2. Prediction results of larch wood density based on different pretreatment methods

View table

View in Article

Table 2. Prediction results of larch wood density based on different pretreatment methods

预处理方法	主因子数	校正集		预测集
预处理方法	主因子数	R²	RMSEC	R²	RMSEP
none	8	0.476 6	0.048 0	0.587 1	0.042 2
MSC	7	0.476 2	0.048 1	0.591 1	0.042 1
SNV	7	0.480 7	0.043 2	0.652 7	0.038 8
SNV+DT	6	0.495 5	0.044 1	0.711 0	0.037 4
MAS	8	0.554 9	0.040 0	0.701 6	0.039 5
SGS	8	0.548 5	0.042 3	0.699 4	0.040 1
MSC+MC+Auto	5	0.525 7	0.048 8	0.689 2	0.034 9
SNV+DT+MC+Auto	5	0.534 9	0.045 3	0.721 1	0.034 7
MAS+MC+Auto	6	0.444 1	0.049 9	0.588 1	0.042 1
SGS+MC+Auto	6	0.423 2	0.050 1	0.587 7	0.042 2

Table 3. Modeling and prediction results of larch wood density based on different outliers elimination methods

View table

View in Article

Table 3. Modeling and prediction results of larch wood density based on different outliers elimination methods

模型	剔除样本数	主因子数	校正集		预测集
模型	剔除样本数	主因子数	R²	RMSECV	R²	RMSEP
Full-PLS	0	12	0.887 5	0.023 1	0.894 2	0.019 6
IFSR-PLS	7	14	0.869 0	0.024 3	0.921 5	0.016 4
MCCV-PLS	2	11	0.882 7	0.023 7	0.894 3	0.019 6
MD-PLS	3	11	0.871 0	0.024 8	0.895 2	0.019 6
HL-PLS	3	11	0.871 0	0.024 8	0.895 2	0.019 6
HLSR-PLS	13	12	0.924 2	0.019 4	0.904 7	0.018 7
SR-PLS	9	12	0.895 3	0.022 5	0.849 8	0.023 0
ODXY-PLS	4	12	0.875 5	0.024 7	0.908 1	0.018 3

Table 4. Modeling and prediction results of larch wood density based on PSO-SVR, BPNN and PLS methods
View table
View in Article
Table 4. Modeling and prediction results of larch wood density based on PSO-SVR, BPNN and PLS methods
模型校正集预测集
R² RMSECV R² RMSEP
PSO-SVR 0.993 3 0.005 5 0.932 1 0.015 4
PLS 0.869 0 0.024 3 0.921 5 0.016 4
BPNN 0.964 5 0.012 7 0.913 1 0.017 7

Tools

Get Citation

Copy Citation Text

Zhe-yu ZHANG, Yao-xiang LI, Zhi-yuan WANG, Chun-xu LI. NIR Model Optimization Study of Larch Wood Density Based on IFSR Abnormal Sample Elimination[J]. Spectroscopy and Spectral Analysis, 2022, 42(11): 3395

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Research Articles

Received: Sep. 23, 2021

Accepted: --

Published Online: Nov. 23, 2022

The Author Email: ZHANG Zhe-yu (1453029789@qq.com)

DOI:10.3964/j.issn.1000-0593(2022)11-3395-08

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Statistical analysis of correction set and prediction set results (g·cm-3)

Table 1. Statistical analysis of correction set and prediction set results (g·cm-3)

Table 2. Prediction results of larch wood density based on different pretreatment methods

Table 2. Prediction results of larch wood density based on different pretreatment methods

Table 3. Modeling and prediction results of larch wood density based on different outliers elimination methods

Table 3. Modeling and prediction results of larch wood density based on different outliers elimination methods

Table 4. Modeling and prediction results of larch wood density based on PSO-SVR, BPNN and PLS methods

Table 4. Modeling and prediction results of larch wood density based on PSO-SVR, BPNN and PLS methods

Table 1. Statistical analysis of correction set and prediction set results (g·cm^-3)

Table 1. Statistical analysis of correction set and prediction set results (g·cm^-3)