Potato Plant Leaf Disease Detection and Recognition Using R-CNN Model

A smart farming system that makes use of the appropriate infrastructure is an example of a new technology that aids in the improvement of the quality and quantity of the country's agricultural goods, such as potatoes. Diseases cannot be prevented since developing potato plants requires consideration of factors such as the environment, soil, and quantity of sunlight. Deep learning has resulted in recent advances in computer vision. These advancements have enabled Potato to utilize a camera to diagnose disease. This study developed an effective new method for detecting infections in potato plants. Yukon Gold, a kind of potato, served as the trial subject. Specifically, Alternaria, Blackleg, and Target Spot were targeted for detection using this method. Using 5,615 images of diseased and healthy Potato plant leaves taken in controlled environments, deep CNN is trained to identify between the presence and absence of three illnesses. To identify the Potato illnesses affecting the tracked plants, the system used a CNN. Anomaly detection using the F-RCNN-trained model achieved a confidence score of 80%, while illness identification using the Transfer Learning model achieved an accuracy of 97.85%. The effectiveness of the automated image-capturing system for diagnosing diseases in Potato plant leaves was determined to be 93.33%.