Plant Disease Identification Using Convolution Neural Networks
Dipra Mitra, Shikha Gupta, Durgesh Srivastava, Sudeshna Sani
Abstract
The most extensively utilized and comprehensive common-sense cognitive engine in the ecosphere is artificial intelligence (AI). The cloud SaaS model and the concept of AI business platforms are practically affluent. It is about AI technologies that can work with other digital systems, so one of the most efficient strategies to protect plants from illnesses in a complicated environment is to detect infections early on. Plant disease detection has become digitized and data-driven as smart farming has expanded, which has the potential to make decisions automatically and intelligently, as well as smart analytics and management, thanks to continual improvements in the field of computer vision. AI-based machine learning (ML) and deep learning (DL) models have recently made substantial progress in the field of digital image processing. Researchers are eager to learn more about how to use a good ML or DL model to identify plant diseases, when pests attack plants and crops, it has an impact on the country’s agricultural output. Usually, farmers or professionals use their naked eyes to detect and identify diseases in plants. However, this procedure can be time-consuming, costly, and imprecise. Automatic detection using image processing techniques yields rapid and reliable results. Based on leaf image classification and deep convolutional networks, this research investigates a unique strategy for constructing a plant disease identification model. Computer vision advances have the potential to widen and improve precision plant protection while also growing the market for computer vision applications in precision agriculture. The novel training method and the technique employed allow for a rapid and painless installation of the system in practice. In this chapter, furthermore, the life cycle of the automatic plant disease detection approach which includes data collection, segmentation, feature extraction, and classification is also mentioned.