Deep Learning for Real-Time Leaf Disease Detection: Revolutionizing Apple Orchard Health

Deep learning approach, particularly integrated neural architecture for real-time Single Shot multibox detector (INAR-SSD), is the most effective approach in the context of fast detection and classification of apple leaf diseases in apple orchards. Advanced data augmentation and complex image annotation techniques are using in the development of apple leaf diseases detection dataset (ALDD), which is the main purpose of this study. Apple leaf diseases detection dataset plays a vital role in the field of identifying apple leaf disease which is the key challenge of this study. The novel INAR-SSD model, created by the combination of SSD architecture with an integrated neural architecture, is proposed in this research. Utilizing the various functionalities of GoogLeNet, Inception Structure and Rainbow concentration the innovative INAR-SSD model is trained to identify various types of apple leaf diseases that effect on apple leaves. To categorize the diseases precisely, the INAR-SSD model follows digital image processing such as noise removing, overfitting reducing, feature extraction etc. Another purpose of study is to report the findings from the experiments that evaluate the performance of the proposed model. The INAR-SSD model successfully detects the illnesses in apple leaves with an accuracy of 97%, evidenced by its mean Average Precision(mAP), and the processing speed of 23.13 Frames Per Second (FPS).