Optimizing Convolutional Neural Networks: A Comparative Study of Gradient-Descent, Adam, and RMSprop Optimizers for Accuracy and Loss in Apple Leaf Disease Detection

This paper investigates convolutional neural network (CNN) optimization algorithms for apple leaf disease detection by contrasting the accuracy and loss metrics of Gradient Descent, Adam, and RMSprop optimizers. The efficiency of each optimizer is assessed using a dataset of 480 scaled leaf photos that show illnesses such as scab, cedar rust, and black rot. Adam achieved the highest accuracy of 93.20 percent, followed by Gradient Descent at 83.29 percent and RMSprop at 33.59 percent. The findings show clear performance disparities. These results contribute to sustainable farming practices by providing important insights into the best optimization techniques to choose for CNN-based disease detection tasks in precision agriculture.