Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Determination of oil content in peanuts based on near infrared spectroscopy

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  2021 ASABE Annual International Virtual Meeting  2100296.(doi:10.13031/aim.202100296)
Authors:   Baoqiong Dai, Yankun Peng, Yali Wang
Keywords:   Multiple peanuts, near-infrared spectroscopy, oil content

Abstract. Peanut is one of the most important oil crops in the world, and its oil content is an important indicator reflecting the quality of peanuts. This paper built a near-infrared spectroscopy acquisition system by itself to explore the feasibility of using near-infrared reflectance spectroscopy combined with chemometric methods to establish a peanut seed oil content detection model. A total of 45 peanut samples of different varieties were selected in the experiment, and each peanut sample was about 70g. First, the spectra (980 nm~1700 nm) of the peanut samples were collected, and six preprocessing methods were used to preprocess the original spectra. Then, the competitive adaptive reweighting sampling (CARS) and successive projection algorithm (SPA) were used to extracted characteristic wavelengths, and established partial least square regression (PLSR) peanut oil prediction model based on the full wavelengths and characteristic wavelengths respectively. The comparison results show that the modeling effect after Savitzky-Golay smoothing combined with first-order derivation (SG-FD) preprocessing is better than that of the original spectrum and other preprocessing methods; the CARS-PLSR model based on characteristic wavelengths had the best prediction ability. The correlation coefficient (Rp) and root mean square error (RMSEP) of the verification set were 0.9264 and 1.0119 %, respectively. The research results show that the self-built near-infrared reflection collection system is feasible to predict the oil content of multiple grain peanuts. This research provides a theoretical basis for high-efficiency, non-destructive, and low-cost detection of oil content in multiple grain peanuts.

(Download PDF)    (Export to EndNotes)