Fast Determination of the Rubber Content in Taraxacum kok-saghyz Fresh Biomass Using Portable Near-Infrared Spectroscopy and Pyrolysis–Gas Chromatography

Chen, Ying; Gao, Shunkai; Dong, Yiyang; Ma, Xiang; Li, Jia-Ru; Guo, Minmin; Zhang, Jiaheng

doi:10.1007/s41664-022-00217-z

Cited by 9 publications

(9 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SNV can reduce interference caused by physical differences in samples [ 32 , 33 ], and MSC can eliminate wavelength shifts caused by sample scattering [ 34 ]. Derivative is a commonly used spectral preprocessing method in the establishment of rubber content prediction models [ 22 – 26 ], FD algorithm has the advantage of eliminating baseline drift and stacking effects, improving spectral resolution, and effectively removing interference from constant baselines and backgrounds [ 35 ].…”

Section: Methodsmentioning

confidence: 99%

“…However, due to environmental and other factors, there exist some noise bands in the near-infrared spectrum which can hinder the predictive performance of the model. To address this issue, the competitive adaptive reweighted sampling (CARS) method [ 37 ] and the previously discovered characteristic bands of NR of TKS [ 26 ] were employed to screen the spectra and reduce the dimensionality of the data, thus reducing the computational complexity and partial noise of the model and minimizing the risk of overfitting. The constrained algorithm for regression variable selection (CARS) is a method that combines MCS with the regression coefficients of partial least squares (PLS) model for feature variable selection, mimicking the principle of “survival of the fittest” from Darwin’s theory [ 37 ].…”

Section: Methodsmentioning

confidence: 99%

“…It resolves the issue of having more samples than variables in multiple linear regression models and is effective when the variables are highly linearly correlated. It has been used to construct an NR content prediction model for TKS fresh roots [ 26 ], displaying impressive predictive performance.…”

Section: Methodsmentioning

confidence: 99%

“…These findings confirm that preprocessing methods like smoothing, SNV, and derivatives can effectively remove some environmental errors in the spectra and enhance spectral features related to NR content. In 2022, Chen et al [ 26 ] discovered that the NIR spectral range of the TKS roots contains characteristic bands with higher resolution of NR and successfully established a PLS prediction model for the NR content in fresh TKS roots, with a predictive set R 2 of 0.97. However, there are no reports in the literature on NR content prediction models for dry TKS roots.…”

Section: Introductionmentioning

confidence: 99%

“…This approach includes smoothing, SNV, and derivative processing, which have been previously employed in rubber content prediction models [ 22 , 24 ], and MSC, a method similar to SNV, frequently used in the establishment of spectral quantitative models [ 27 ]. Current rubber content prediction models are predominantly linear, with PLS being the sole modeling algorithm applied in previous studies [ 22 – 26 ]. Therefore, in addition to PLS, this study incorporates three nonlinear modeling algorithms commonly used in quantitative model establishment: random forest (RF), lightweight gradient augmentation machine learning (LightGBM), and convolutional neural network (CNN), for comparison.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A prediction model of rubber content in the dried root of Taraxacum kok-saghyz Rodin based on near-infrared spectroscopy

Chen,

Yan,

Tuoheti

et al. 2024

Plant Methods

View full text Add to dashboard Cite

Background Taraxacum kok-saghyz Rodin (TKS) is a highly potential source of natural rubber (NR) due to its wide range of suitable planting areas, strong adaptability, and suitability for mechanized planting and harvesting. However, current methods for detecting NR content are relatively cumbersome, necessitating the development of a rapid detection model. This study used near-infrared spectroscopy technology to establish a rapid detection model for NR content in TKS root segments and powder samples. The K445 strain at different growth stages within a year and 129 TKS samples hybridized with dandelion were used to obtain their near-infrared spectral data. The rubber content in the root of the samples was detected using the alkaline boiling method. The Monte Carlo sampling method (MCS) was used to filter abnormal data from the root segments of TKS and powder samples, respectively. The SPXY algorithm was used to divide the training set and validation set in a 3:1 ratio. The original spectrum was preprocessed using moving window smoothing (MWS), standard normalized variate (SNV), multiplicative scatter correction (MSC), and first derivative (FD) algorithms. The competitive adaptive reweighted sampling (CARS) algorithm and the corresponding chemical characteristic bands of NR were used to screen the bands. Partial least squares (PLS), random forest (RF), Lightweight gradient augmentation machine (LightGBM), and convolutional neural network (CNN) algorithms were employed to establish a model using the optimal spectral processing method for three different bands: full band, CARS algorithm, and chemical characteristic bands corresponding to NR. The model with the best predictive performance for high rubber content intervals (rubber content > 15%) was identified. Result The results indicated that the optimal rubber content prediction models for TKS root segments and powder samples were MWS–FD CASR–RF and MWS–FD chemical characteristic band RF, respectively. Their respective $${\text{R}}_{{\text{P}}}^{2}$$ R P 2 , RMSEP, and RPDP values were 0.951, 0.979, 1.814, 1.133, 4.498, and 6.845. In the high rubber content range, the model based on the LightGBM algorithm had the best prediction performance, with the RMSEP of the root segments and powder samples being 0.752 and 0.918, respectively. Conclusions This research indicates that dried TKS root powder samples are more appropriate for constructing a rubber content prediction model than segmented samples, and the predictive capability of root powder samples is superior to that of root segmented samples. Especially in the elevated rubber content range, the model formulated using the LightGBM algorithm has superior predictive performance, which could offer a theoretical basis for the rapid detection technology of TKS content in the future.

show abstract