Orange has abundant nutritional properties and is consumed worldwide. Sorting oranges of different masses based\r\non their physical traits could help reduce packaging and transportation cost. The ââ?¬Ë?Bloodââ?¬â?¢ cultivar of Iranian oranges from\r\nKermanshah province of Iran (7.03 Ã?°E 4.22 Ã?°N) was used in this study. 100 samples were randomly selected. During the\r\ntwo-day experiment, all measurements were carried out inside the laboratory at mean temperature of 24Ã?°C. In this study, some\r\nphysical properties of ââ?¬Ë?Bloodââ?¬â?¢ orange were measured, such as length, width, thickness, volume, mass, mean value of geometric\r\ndiameter, sphericity and projected area. ANFIS and linear regression models were employed to predict the mass based on\r\nsphericity and mean of projected area inputs. In ANFIS model, samples were divided into two sets, with 70% for training set\r\nand 30% for testing set. The coefficient of determination (R2) for ANFIS and linear regression models were 0.983 and 0.927,\r\nrespectively. It is shown that the mass can be estimated based on ANFIS model better than linear regression model.
Loading....