Leaf Disease Prediction Using Transfer Learning

Leaf Disease Prediction Using Transfer Learning

Sonal Mangal Baviskar

Prof. Pratik Buchke, Associate Professor

Department of Software Engineering

Oriental Institute of Science and Technology, Bhopal

ABSTRACT

There are some most extensively planted vegetable crop in India's agricultural lands. Although the tropical environment is favourable for its growth, specific climatic conditions and other variables influence  plant growth. In addition to these environmental circumstances and natural disasters, plant disease is a severe agricultural production issue that results in economic loss. Therefore, early illness detection can provide better outcomes than current detection algorithms. As a result, deep learning approaches based on computer vision might be used to detect diseases early. This study thoroughly examines the disease categorization and detection strategies used to identify plant leaf diseases. The pros and limitations of the approaches provided are also discussed in this study. Finally, employing hybrid deep-learning architecture, this research provides an early disease detection approach for detecting tomato leaf disease.

INDEX TERMS Artificial Intelligence, Convolutional Neural Network (CNN), deep learning, leaf disease, tomato leaf , multiclass classification.