Abstract

The aim of this study is to estimate the solid density of cane stalk using the short-wave near infrared spectroscopy across mobile near infrared spectrometer with the wavelength of 570–1031 nm. The idea was to follow up the change of individual sugarcane stalk weight using non-destructive method, then the knowledge of solid density was required. The sugarcane stalk weight could be calculated by multiplication between solid density and its volume, which could be calculated from cane stalk diameters and its length 40 cm.The model was optimized using PLS regression. To get the effective model, the effect of waxy cover, different integration time, different wavelength range, and various pre-processing technique were considered. The result showed that original sample model gave poor performance because of the effect of waxy cover. Meanwhile, NIR spectra without waxy cover model provided a good performance. The best model was achieved where the NIR spectra scanned with integration time of 200 ms, wavelength of 700–1000 nm, and with smoothing moving average were applied. The standard error of prediction (SEP), relative standard error of prediction (RSEP) were 22.131 kg/m3 and 2.000%, respectively. It could be applied for screening. Hence the outcome showed that NIR spectroscopy could be used as a non-destructive technique to follow up the mass change of individual cane stalk. For cane breeding and mass yield programs, the knowledge could be used to estimate the growing rate when the environmental condition was changed. The model would benefit to breeding with time and cost consumption reduction.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call