Across different instruments about tobacco quantitative analysis model of NIR spectroscopy based on transfer learning.

With the development of near-infrared (NIR) spectroscopy, various calibration transfer algorithms have been proposed, but such algorithms are often based on the same distribution of samples. In machine learning, calibration transfer between types of samples can be achieved using transfer learning and does not need many samples. This paper proposed an instance transfer learning algorithm based on boosted weighted extreme learning machine (weighted ELM) to construct NIR quantitative analysis models based on different instruments for tobacco in practical production. The support vector machine (SVM), weighted ELM, and weighted ELM-AdaBoost models were compared after the spectral data were preprocessed by standard normal variate (SNV) and principal component analysis (PCA), and then the weighted ELM-TrAdaBoost model was built using data from the other domain to realize the transfer from different source domains to the target domain. The coefficient of determination of prediction (R 2) of the weighted ELM-TrAdaBoost model of four target components (nicotine, Cl, K, and total nitrogen) reached 0.9426, 0.8147, 0.7548, and 0.6980. The results demonstrated the superiority of ensemble learning and the source domain samples for model construction, improving the models' generalization ability and prediction performance. This is not a bad approach when modeling with small sample sizes and has the advantage of fast learning.

Customized optimization of lignocellulolytic enzyme cocktails for efficient conversion of pectin-rich biomass residues.

Pectin is a major component in many agricultural feedstocks. Despite the wide use in industrial production of cellulases and hemicellulases, the fungus Trichoderma reesei lacks a complete enzyme set for pectin degradation. In this study, three representative pectinolytic enzymes were expressed and screened for their abilities to improve the efficiency of T. reesei enzymes on the conversion of different agricultural residues. By replacing 5 % of the T. reesei proteins, endopolygalacturonase and pectin lyase remarkably increased the release of sugars from inferior tobacco leaves. In contrast, pectin methylesterase showed the strongest improving effect (by 31.1 %) on the hydrolysis of beetroot residue. The pectin in beetroot residue was only mildly degraded with the supplementation of pectin methylesterase, which allowed the extraction of pectin keeping the original emulsifying activity with a 51.1 % higher yield. The results provide a basis for precise optimization of lignocellulolytic enzyme systems for targeted valorization of pectin-rich agricultural residues.

Quality evaluation of flue-cured tobacco by near infrared spectroscopy and spectral similarity method.

Herein we propose near infrared (NIR) spectroscopy as a rapid method of evaluating the quality of agricultural products. Unlike existing quantitative or qualitative models, quality similarity is characterised using spectral similarity. Key factors of the spectral similarity method were investigated, including variable selection, pre-processing and similarity measures. Sophisticated techniques were developed to ensure the reliability of similarity algorithm. The proposed method was tested by quality similarity of flue-cured tobacco samples. The results demonstrated that the quality-related factors between the target and the similar samples (determined by spectral similarity), showed high similarities. This new method has the potential to characterise product quality effectively and could be a useful new alternative to the widely used PLS models.