Development of Efficient CNN models for plant disease identification

Abstract

Automatic identification of plant disease from leaf images has been a subject of interest for more than two decades. A number of Machine Learning (ML) algorithms and Convolution Neural Network (CNN) models have been proposed for identification of various crop diseases. CNN models are based on Deep Learning Neural Networks and differ inherently from traditional Machine Learning algorithms like k-NN, Decision-Trees etc. Moreover, the performance of Deep Neural Network based approaches are better as compared to traditional Machine Learning approaches as these models extract the features from training data automatically. In past, the researchers have proposed many CNN architectures such as VGG16, VGG19, InceptionV3, MobileNet, ResNet50, etc. for the classification of 1000 class imagenet dataset. These models can also be utilized for the classification of other data sets by transfer learning. While pre-trained CNN models perform fairly well, they tend to be computationally heavy due to large number of parameters involved.