Moisture Content Quantization of Masson Pine Seedling Leaf Based on Stacked Autoencoder with Near-Infrared Spectroscopy

• Published in: Journal of electrical and computer engineering Vol. 2018; no. 2018; pp. 1 - 8
• Main Authors: Ni, Chao, Wang, Dongyi, Zhang, Yun
• Format: Journal Article
• Language: English
• Published: Cairo, Egypt Hindawi Publishing Corporation 01-01-2018; Hindawi; Hindawi Limited
• Subjects: Artificial intelligence; Biodiesel fuels; Calibration; Cost analysis; Data analysis; Feature extraction; Genetic algorithms; Informatics; Infrared analysis; Infrared spectra; Infrared spectroscopy; International conferences; Moisture content; Near infrared radiation; Pine; Predictions; Regression analysis; Regression models; Signal processing; Spectroscopic analysis; Spectrum analysis; Support vector machines; China
• ISSN: 2090-0147; 2090-0155
• DOI: 10.1155/2018/8696202

Masson pine is widely planted in southern China, and moisture content of the pine seedling leaves is an important index for evaluating the vigor of seedlings. For precisely predicting leaf moisture content, near-infrared spectroscopy analysis is applied in the experiment, which is a cost-effective, high-speed, and noninvasive material content prediction tool. To further improve the spectroscopy analysis accuracy, in this study, a new analysis model is proposed which integrates a stacked autoencoder for extracting hierarchical output-related features layer by layer and a support vector regression model to leverage these features for precisely predicting moisture contents. Compared with traditional spectroscopy analysis method like partial least squares regression and basic support vector regression, the proposed model shows great superiority for leaf moisture content prediction, with R2 value 0.9946 and root-mean squared error (RMSE) value 0.1636 in calibration set and R2 value 0.9621 and RMSE 0.4249 in prediction set.