Comparative Analysis of ResNet based Plant Disease Detection for Agricultural Applications

Plant diseases arise from causative agents that induce abnormal physiological processes, disrupting a plant’s typical growth, structure, function, and other essential activities. Infectious plant diseases stem from pathogenic organisms, including fungi, bacteria, mycoplasma, viruses, and parasitic flowering plants. These organisms have the capability to reproduce and spread, posing a threat to the health of the plant. Identification of plant diseases involves recognizing abnormal growth patterns or signs of disease-causing organisms, such as bacterial slime or insect larvae. Early detection is crucial, as it enables prompt removal or treatment of the affected plants, preventing the further spread of the disease to neighboring plants. The study and detection of plant diseases constitute essential research areas with various benefits. Monitoring extensive crop fields allows for the timely identification of disease symptoms as soon as they manifest. Taking immediate action upon detecting these symptoms is crucial to prevent widespread damage. Research in this field contributes to the development of effective strategies for disease management in agriculture, ensuring the health and productivity of crops. This study gives an analysis of some techniques, systems used for identification of plant disease with emphasis on deep learning methods of Convolution Neural Network (CNN).