Artificial neural network and regression model for estimating the volume of apple

Physical properties of agricultural products are the most important parameters that influence grading and packaging systems. Volume of agricultural products is one of these attributes which should be measured accurately. The mathematical regression and two-hidden-layer feed-forward artificial neural network were used to develop models for estimating volume of apple. The input parameters of the network were major diameter, intermediate diameter, minor diameter and mass while the output of the network was volume. Correlation between dimensions and volume by using regression model is 0.97. Also Paired t-test and Bland-Altman approach indicated that the difference between the real volume and the volume obtained by ANN method was not significantly different (p<0.05) and mean difference between them was resulted to be 2.38 cm3.This study approves that ANN method can predict the volumes of agricultural products effectively

Keywords:

Artificial neural network , Fruit , Post-harvest , Regression model , Volume