Prediction and classification of sugar content of sugarcane based on skin scanning using visible and shortwave near infrared

• Published in: Biosystems engineering Vol. 115; no. 2; pp. 154 - 161
• Main Authors: Mat Nawi, Nazmi, Chen, Guangnan, Jensen, Troy, Mehdizadeh, Saman Abdanan
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.06.2013; Elsevier
• Subjects: Agricultural machinery and engineering; Agronomy. Soil science and plant productions; Biological and medical sciences; engineering; Fundamental and applied biological sciences. Psychology; Generalities. Biometrics, experimentation. Remote sensing; least squares; neural networks; prediction; reflectance; spectroscopy; sugar content; sugarcane; total soluble solids; Bioengineering; Monocotyledones; Vegetals; Biochemical compound; Prediction; Tropical crop; C4-Type; Electromagnetic wave; Sugar plant; Short wave; Near infrared radiation; Engineering; Saccharum officinarum; Gramineae; Content; Angiospermae; Classification; Spermatophyta; Agriculture; Sugar cane; Carbohydrate; Skin; Sugar
• ISSN: 1537-5110; 1537-5129
• DOI: 10.1016/j.biosystemseng.2013.03.005

The potential application of a visible and shortwave near infrared (Vis/SWNIR) spectroscopic technique as a low cost alternative to predict sugar content based on skin scanning was evaluated. Two hundred and ninety one internode samples representing three different commercial sugarcane varieties were used. Each sample was scanned at four scanning points to obtain the spectra data which was later correlated with its °Brix (soluble solids content) values. Partial least square (PLS) model was developed and applied to both calibration and prediction samples. Using reflectance spectra data, the model had a coefficient of determination (R2) of 0.91 and root means square error of predictions (RMSEP) of 0.721 °Brix. The artificial neural network (ANN) was also applied to classify spectra data into five °Brix categories. The ANN has yielded good classification performance, ranging from 50 to 100% accuracy with an average accuracy of 83.1%. These results demonstrated that the Vis/SWNIR spectroscopy technique could be applied to predict sugarcane °Brix in the field based skin scanning method. ► Application of Vis/SWNIR to predict sugar content based on skin scanning explored. ► Partial least square (PLS) model applied to calibration and prediction samples. ► Model R2 of 0.91 and RMSEP of 0.721 °Brix. ► ANN used to classify spectra data into five °Brix categories yielded good classification. ► Vis/SWNIR field based skin scanning method can be applied to predict sugarcane °Brix.