Abstract

In a country like India, where a major portion of the population is involved in agriculture, it is crucial to detect plant diseases at early stages. Early disease detection is important for better yield and quality of crops. Reduction in the quality of agricultural products due to diseased plants can lead to huge economic losses for individual farmers. Various methods for detecting plant diseases have been developed as a means to ensure food security and reduce food waste through early detection. Precision farming has been developed through these technological advancements, and the application of machine learning is steadily gaining popularity in this industry as a means of solving this problem. Faster and precise prediction of plant disease could help in reducing the losses. Significant advancements and developments in deep learning have provided the opportunity to improve the performance and accuracy of detection of objects and recognition systems. This paper focuses on one of the major growing agricultural crop in India."CornCare: Plant Disease Defender" is a groundbreaking project aimed at revolutionizing agriculture, particularly in the context of maize cultivation, a vital crop worldwide. Leveraging cutting-edge Convolutional Neural Networks (CNNs) technology, the project focuses on early disease identification and categorization in corn plants. The increasing global population has heightened food security concerns, making corn, with its diverse applications in human and animal diets and various industries, a crucial resource. Although there is a wide variety of corn plant diseases, this paper covers just three major ones from the Kaggle database: Common rust, Northern leaf blight and Grey leaf spot. These diseases pose significant threats, often going unnoticed until they cause substantial crop losses and endanger food supplies. "CornCare" addresses these challenges by reducing reliance on visual inspections and the expertise of plant pathologists, thereby enhancing the efficiency and accessibility of disease identification. This project holds immense importance for the agriculture industry, ensuring early diagnosis and treatment of diseases to improve food security, stabilize markets, and promote sustainable agriculture practices. By encouraging responsible agrochemical use and reducing the environmental impact of corn farming, "CornCare" contributes to economic stability and agricultural development. To determine the model's effectiveness, it must first be trained, then validated, and ultimately tested. In testing, the proposed algorithm achieved an efficiency rate of 92%, showcasing the dataset's efficacy in accurately identifying and categorizing corn plant diseases. This result highlights the potential of the "CornCare" project to significantly improve disease management in maize crops and further support sustainable agriculture practices.