Application of Artificial Neural Networks (ANN) and Image Processing for Prediction of Gravimetrical Properties of Roasted Pistachio Nuts and Kernels

Abstract

Roasting is among the most common methods of nut processing causing physical and chemical changes and ultimately increasing overall acceptance of the product. In this research, the effects of temperature (90, 120 ,and 150°C), time (20, 35 ,and 50 min) ,and roasting air velocity (0.5, 1.5 ,and 2.5 m/s) on gravimetrical properties of pistachio nuts and kernels including unit mass, true density, ounce, uniformity, size and shell percentage were investigated. Gravitational characteristics were measured by experimental and image processing methods. Artificial neural network method was used to predict the relationship between characteristics obtained from experiments and image processing. Volume, unit mass and true density for pistachio nuts were in a range of 1.06 – 1.24 mm³, 0.92 – 1.08 g and 866.01 - 871.35 kg / m³, respectively and they were0.61-0.77 mm³, 0.53 - 0.67 g and 862.21 - 871.29 kg / m³ for pistachio kernels. Number of pistachio nuts was found to be 29-32 per ounce and 102-109 per 100 grams. Uniformity of pistachios was in a range of 1.24-1.50 and their average kernel ratio was higher than 50%. Thus, it can be said that, they were of superior quality. Shell percentage of pistachio nuts was in a range of 38.24–41.98%. Results of the study revealed that ,artificial neural network could properly predict volume and mass of pistachio nuts, but, it had not appropriate ability to predict apparent density.