Revolutionizing Peach Farming: Advanced Disease Classification Through CNN And Random Forest

Author: Madhavi Tripathi & Himani Tyagi

DOI Link ::   https://doi-ds.org/doilink/05.2024-49767638/BIJMRD/Vol -1 / 2/Jan-2024/A4

Abstract. In India, agriculture is the main source of income. Its contribution to the Gross Domestic Product is noteworthy. Crop diseases, however, have a major effect on food security and agricultural output. For efficient disease control, illnesses in plants or crops must be accurately and promptly classified. End human trafficking, achieve food security and improved nutrition, and promote sustainable agriculture” is something that these realities motivate our team to focus on. The team decided to work on “Peach Leaf Disease Classification” after conducting study. Convolutional neural networks and Random Forest algorithm used were used to do the comparative study of the Peach leaf disease classification as the conclusion of the literature review. Plant disease classification and identification are being done more and more with machine learning models. As a result, our group divided up four distinct architectures among themselves. Sequential CNN reported accuracy of 96%. AlexNet provided 99.25% accuracy, VGG provided 97.50%, while ResNet-50 provided 99.62% accuracy. Our group is prepared to concentrate on creating deep learning models that can reliably identify and categorize a wide variety of subcategories inside more general classes.

Keywords: Peach leaf, Random Forest,Bacterial spot, ResNet-50, CNN, VGG, AlexNet, Binary Classification