ABSTRACT
The production of agricultural products is essential to human survival. Agricultural goods are vital to food security and economic development, despite being susceptible to numerous bacterial, viral, and fungal diseases. Apple disease and insect infestation are the main causes of reduced fruit yield. Apples share both characteristics. As a result of low apple production, the apple industry loses money every year. Rare plant diseases are diagnosed by botanists and other agricultural specialists, resulting in excessive pesticide use. Early detection of apple diseases can help reduce the spread of diseases, enhance productivity, and eventually eliminate them. Several factors can make diagnosing and identifying early illness difficult, including the presence of several symptoms on a single leaf. These factors include an uneven backdrop, leaf color changes caused by infected cells’ age, and the size of disease spots. Using machine learning algorithms and sophisticated machine learning techniques, this study constructs a robust model for accurate diagnosis and classification of apple leaf disease. Plant loss will be reduced and detection accuracy will be improved. For detecting objects and diagnosing apple leaf disease, GLCM, SVM, and deep learning algorithms are recommended. As a result, early detection and accurate diagnosis will be possible. In the future, the system may be able to diagnose leaf diseases in many plants in real time. Consequently, computer science and agriculture will become more accurate and reliable, as apple tree diseases, black rot, fish scale disease, and snow apple rust are being investigated.
