Machine Intelligence, Big Data Analytics, and IoT in Image Processing

5Plant Pathology Detection Using Deep Learning

Sangeeta V.¹, Appala S. Muttipati²* and Brahmaji Godi³

¹ Anil Neerukonda Institute of Technology and Sciences, Department of Computer Science and Engineering, Visakhapatnam, Andhra Pradesh, India

² iNurture Education Soultions Pvt. Ltd, Department of IT Academics, Bangalore, India

³ Raghu Institute of Technology, Department of Computer Science and Engineering, Visakhapatnam, Andhra Pradesh, India

Abstract

As plant diseases are a major threat to developing countries like India, agriculture plays a significant or critical role in the economy. The prognosis of plant diseases at the very beginning reduces the risk to substance security. Biological examination of plants or visual inspection of plants by experts are expensive and have a lot of delays. This has paved a path to implement computer methodologies to detect diseases and suggest pesticides. Latest technologies like image processing and computer vision have developed many algorithms for early detection. Advances in computer vision are facilitated by deep learning (DL) for smart phone-based diagnosis systems. DL techniques use convolution neural networks with familiar residual neural network architecture, which is suitable to develop a model for early disease detection. The model was developed using a dataset of 7,000 healthy and unhealthy plant leaf images. The results were encouraging and showed 95% accuracy in comparison with existing methods.

Keywords: Plant leaf, leaf disease, ...