Location: Peanut Research
Title: Nondestructive moisture content determination of valencia peanuts by NIR reflectance spectroscopic method Authors
|Govindarajan, K -|
|Subbiah, J -|
Submitted to: Proceedings of the American Society of Agricultural and Biological Engineers International (ASABE)
Publication Type: Proceedings
Publication Acceptance Date: December 2, 2009
Publication Date: June 20, 2010
Citation: Kandala, C., Sundaram, J., Naganathan, K.G., Subbiah, J. 2010. Nondestructive Moisture Content Determination of Valencia Peanuts by NIR Reflectance Spectroscopic Method. Proceedings of the American Society of Agricultural and Biological Engineers International (ASABE). Interpretive Summary: Presently available commercial moisture meters can determine the MC of shelled peanuts only. It would be convenient if the MC of peanuts could be determined rapidly and nondestructively at the time of purchase and drying. Ability to determine the MC of in-shell peanuts would save lot of time and labor during these processes. In this work an indigenously developed NIR reflectance spectrometer was used to estimate the MC of both in-shell and shelled peanuts. The process is rapid. The results are very encouraging and this method could be used to develop a commercial NIR reflectance spectrometer that could rapidly and nondestructively measure MC of in-shell peanuts.
Technical Abstract: In earlier work, the possibility of determining moisture content (MC) nondestructively in in-shell peanuts was investigated by the authors, using a commercial NIR Spectrometer. It is well known that with most of the commercial moisture meters presently available, MC determination involves shelling and cleaning of the peanut samples, and in some cases some sort of sample preparation such as grinding. This is cumbersome, time consuming and destructive. If MC of the peanuts could be measured on the in-shell peanuts itself rapidly and nondestructively, it would be useful for the peanut industry. In the work presented here, an indigenously developed NIR spectrometer was used to make reflectance measurements on Valencia type peanuts in both un-shelled and shelled forms. The MC range of the peanuts tested was between 6% and 24%. NIR reflectance measurements were made at 1 nm intervals in the wavelength range of 1000 nm to 25000 nm and the spectral data was modeled using partial least squares regression (PLSR). Eight different models were developed by utilizing different data preprocessing methods such as Norris-Gap first derivative with a gap size of 3, peak normalization with 1680 nm (which is the no absorbance wavelength for water), and absorbance transformation. From these, a suitable model was selected based on model fitness measures. Predicted values of the samples tested in the above range were compared with the values determined by the standard air-oven method. The predicted values agreed well with the air-oven values with an R2 value of 0.91 and a standard error of prediction (SEP) of 1.37 for averaged spectra and an R2 value of 0.88 and SEP value of 1.55 for individual spectra. Both methods of analysis performed satisfactorily and being rapid, nondestructive and non-contact, may also be suitable for continuous monitoring of MC of in-shell peanuts as they move on conveyor belts during their processing.