Deep Learning Approach for Disease Detection in Sweet Lemon Leaves to Enhance Citrus Fruit Yield

The application of deep learning(DL approaches to the problem of identifying diseases of the sweet lemon leaf is a recent promising outcomes in many artificial intelligence (A) problems. Numerous techniques are used in machine learning (ML) models for classification and detection; however, given the ongoing progress in computer vision, it seems that deep learning (DL) research has a lot of promise for improving accuracy. The presence of diseases affecting sweet lemon leaves has a significant influence on Rayalaseema's economy, leading to lower-quality fruit production and a higher incidence of canker, greening, black spot, and other diseases. There is a shortage of research datasets and a restricted variety of pests, making it challenging to adapt research findings to agriculture in the area of sweet lemon leaf pest disease. The recommended strategy is assessed on the combined sweet lemon leaf disease picture datasets in the proposed study, which offers a method for identifying and categorizing diseases in sweet lemon leaf disease plants using deep learning and image processing. CNN, ResNet, MobileNet, and VGG16 are the models used to identify sweet lemon leaf disease. With an accuracy rate of $98.6 \%$, the suggested MobileNetV2 model outperforms the current CNN technique.