Plants leaf damage detection based on colour and texture using deep learning – AlexNet

Abstract

The primary occupation in India is agriculture. India ranks second in the agricultural output
worldwide. Here in India, farmers cultivate a great diversity of crops. Various factors such as climatic
conditions, soil conditions, various diseases, etc affect the production of the crops. The approach used for the
identification of plant diseases is simply naked eye inspection, which requires more manpower, adequately
equipped laboratories, expensive facilities, etc. The inadequate detection of diseases will lead to inexperienced
use of pesticides, thereby reducing the ability of the crop to fight off the growth of long term pathogen
resistance. The device suggested helps to detect plant disease. For the experiments, the database used is a
dataset for plant village. The classifier is trained using training data and then the performance is correctly
predicted. Convolution Neural Network (CNN) is used which comprises of different layers which are used for
prediction. CNN identifies the plant disease using the colour and texture of the leaf. The proposed system
produced better results when compared to the existing algorithms.