The Implementation of A Crop Diseases APP Based on Deep Transfer Learning

Abstract

Classifying the severity of crop diseases is the staple-basic element of the plant pathology for making disease prevent and control strategies. The diagnosis of disease needs timeliness and accuracy. Thanks to the development and popularity of smart phones and mobile networks, this makes possibly to develop mobile applications that can be widely accepted by users in the agricultural community. This paper provides a system that can detect the severity of crop diseases automatically and intelligently through taking photos. The development of this mobile app is based on deep transfer learning that we proposed an improved method with nearly 92% accuracy based on ResNet 50. The significantly high success rate makes the model a very useful advisory or warning tool. This project provides a new idea and solution for the detection of crop diseases in agriculture. 2020 IEEE.