Determination of Physical Characteristics and Regression Modeling of Orange (Dezful Local Variety) Mass and Volume
Subject Areas :
Keywords: Regression equation, Geometric dimensions, Projected area, Determination coefficient,
Abstract :
In this study, 100 random samples of orange fruit (local variety of Dezful) harvested from a garden and carried to the laboratory and various measurements were done to determine the main physical characteristic of the fruits (including three geometric dimensions, mass, volume and projected area). Analysis of variance of experimental data for the first model showed that there is a significant relationship at 1% level between the mass or volume of orange and its dimensions. Determination coefficients (R2) values of the regression equations for estimating the mass or volume of oranges using three dimensions showed that they can justify 95.97 and 98.01 percent of changes in orange mass or volume, respectively. In the second regression model, the Pearson correlation coefficient between the projected area and mass or volume was 0.979 and 0.966, respectively. This correlation was significant at 1 % level. In comparing the performance of two regression models presented in this study, based on small values for the RMSE and MAPE indices and large value for the EF index, the F (h, w, t) model performance was better than the F (A) model. The determination coefficient (R2) values for the presented regression equations show they have high accuracy in estimating the mass or volume of orange
_||_