Location: Peanut Research
Title: Classification of in-shell peanut kernels nondestructively using VIS/NIR reflectance spectroscopy Authors
Submitted to: Vibrational Spectroscopy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: May 29, 2010
Publication Date: June 30, 2010
Repository URL: http://DOI 0. 1007/s11694-010-9098-9
Citation: Sundaram, J., Kandala, C., Butts, C.L. 2010. Classification of in-shell peanut kernels nondestructively using VIS/NIR reflectance spectroscopy. Vibrational Spectroscopy. 4:82-94. DOI 0. 1007/s11694-010-9098-9. Interpretive Summary: Peanuts produced in United States are considered as high quality peanuts. To continue this quality, grading of farmers stock peanuts should be improved further. When the peanuts are picked from the farmers they are unshelled peanuts. There are some peanuts that contain kernels with damages, immature, discolored kernels etc, which are simply called as damages. Traditional way of grading and counting these types of damages is slow, labor intensive and sometimes inaccurate too. A device which can identify the inferior quality peanut kernels rapidly without shelling them is very useful. Techniques using near infrared (NIR) spectroscopy for food quality measurements are becoming more popular in food processing and quality inspection of agricultural commodities. NIR spectroscopy has several advantages over conventional physical and chemical analytical methods of food quality analysis. It is very rapid and non destructive method and provides more information about the components and its structure present in the food and food products. It also measures more than one parameter simultaneously. . Foss NIR spectroscopy was used to identify the inferior quality kernels without shelling the peanuts. Partial Least Square (PLS) analysis was carried on calibration set and a model was developed to predict the quality using validation set. Using this method the quality of the peanut was predicted with a Standard Error of Prediction (SEP) of 0.401 and Bias of -0.109. Predicted quality character was compared with the actual quality and it was found that 83.3% of the peanuts were predicted well. This method is rapid, non labor intensive and it has the promising application in peanut grading.
Technical Abstract: One of the grading factors for peanuts is their classification into peanuts with good or bad kernels. Traditional manual methods are labor intensive and subjective. A device by which the classification could be done rapidly and without the need to shell the peanuts would be very useful for the peanut industry. In this work VIS/NIR spectroscopy was used for this purpose. Reflectance spectra were collected for peanut pods (in-shell peanuts) in the wavelength range of 400 nm to 2500 nm. A calibration group of about 200 pods were initially scanned to train the computer. Each individual pod was shelled and the kernels were visually examined and classified as bad if they had any kind of damage, discoloration or immaturity. The remaining pods were marked as good ones. Using this classification, principal component analysis models developed could classify over 94% of the pods in the validation group correctly as good or bad. A partial least square model was also developed that had an R2 value of better than 0.90 and classified over 95% of the pods from the validation group as good ones correctly. Both models were found to work very well.