Fruit Recognition and Feature Extraction of Fruits Using Deep Learning

Abstract

Now a days automation in every field becomes common. While we deep dive in the study of nutrients of fruits, it has become necessary to abstract every features for proper intake of nutrients in our day-to-day life. Among these classification of fruits is very essential and challenging task as many fruits looks similar in terms of colour, shape and size. It is a necessity to identify every kind of nutrients which are necessary in our diet just not overtake any nutrients. To identify the type of fruit what it is, is as important as extracting features. Hence Convolutional Neural Network (CNN) based classification method is proposed which gives a better classification result of approximately >92% compared to other proposed methodologies till now. After Detecting which fruit it is a machine learning algorithm is used to give output as nutrients of fruit. Experiments are held with the dataset of more than 4500 fruit images in which various types are fruits images are considered.