An Ensemble-Based Model of Detecting Plant Disease using CNN and Random Forest

The protection of crops, fruits, and vegetables has drawn attention nowadays because of the spreading of diseases in different varieties of crops caused by various pathogens, and become a serious issue in India. This paper will propose a framework for detecting & classifying healthy and unhealthy plant leaves of ten different crops including fruits and vegetables along with various diseases of different crops. The proposed research is an ensemble-based model using a deep learning algorithm i.e., Convolutional Layers of the Convolutional Neural Network, and Random Forest classifier. This proposed framework achieves 99.89 % accuracy during the testing/validation phase on 380000 images taken from standard datasets. Also, the plant disease is detected using RF and CNN separately and then, the accuracy of RF, CNN, and CNN with RF models are compared.