FRUITSCAPE: Fruit disease detection using deep learning

Abstract

Deep Learning is an uprising technology in terms of image classification. It has reached greater heights in forms of performance and accuracy with the advanced ability to perform better in face of large data sets. The traditional approach for model fitting, initiated with the arriving to decision on a model of data followed with a parameter estimated with that data. Machine learning has altered this method completely by using an algorithm to ensure the connection between inputs and outputs instead of starting off with a data model. ML approach in pattern recognition is focused on learning outcomes by perceiving the inputs and detecting critical patterns in it. FruitScape is a disease detection and fertilizer recommendation application built on top of a deep learning module that comprises of the capability to identify different stages the infection level and provide the specific recommendation based on the predictions. It provides an ultimate solution adhering all the problems faced by the farming community due to diseases and inability to identify them in a timely manner. The expertise and experience required by a farmer to predict and decide on a fertilizer to overcome the disease is replaced by fruitscape. This is an escape to the fruit diseases.