Volume Estimation of Non-axisymmetric Fruits and Vegetables using Image Analysis

Volume is one of the most important parameters for fruits and vegetables grading. The conventional methods of volume estimation have some shortcomings in the context of the agricultural industry. It requires a huge number of expert manual resource and a long time to estimate volume for a large number of fruits and vegetable in production. It needs a non-destructive and automated method for volume estimation. The automated volume estimation from image analysis becomes challenging when the fruits and vegetables are irregular and non-axisymmetric in shape. The preprocessed binary version of fruit or vegetable image is split into two parts. One polynomial equation is formed for each part from the valid boundary points. Volume is estimated from those equations. The proposed technique is tested on a dataset of 52 images of potato, which is irregular and non-axisymmetric in shape. The overall correlation coefficient between the conventional technique and the proposed approach for volume estimation is 0.99.