Determination of Volume and Mass of Lemons by Mechine Vision System

Physical characteristics of agricultural products are the most important parameters in the design of grading, conveying, processing and packaging systems. Among these physical characteristics, volume, mass, projected area and center of gravity are the most important for sizing systems. In this study, the segmentation method was used to estimate lemon volume. A total of 50 randomly selected lemons were examined. The mechanized scanning system consisted of two CCD cameras, two capture cards, an appropriate lighting system and a personal computer. The cameras were arranged at right angles to each other to capture perpendicular images of the lemons. The estimated volume using this technique was compared to the actual volume of the lemons, measured by water displacement, using a paired t-test and the Bland-Altman approach. The estimated volume using the mechanized scanning method was not significantly different from that determined by water displacement (p > 0.05). The mean differencebetween water displacement and mechanized scanning was -0.06 cm3. The characterization results of the lemons showed that the computed volume and measured mass parameters were highly correlated, M = 0.8894V + 2.2757, with a coefficient of determination of 0.96. In conclusion, the mechanized scanning technique provides a simple and efficient methodology for estimating lemon volume and mass.