Agriculturalists spend the majority of their financial resources managing and diagnosing plant diseases. Manually identifying plant diseases requires a lot of time and effort. This study recommends deep neural networks to automatically detect and identify plant diseases. The datasets are trained using a Random Forest technique to classify images of diseased and healthy leaves. The suggested ontology is modelled based on the Web Ontology Language (OWL). Support vector machines, convolution neural networks, and artificial neural networks are just a few of the classifiers utilized to extract features. A dataset that was self-gathered during the augmentation phase is used to validate the augmented method as part of the validation process. The majority of machine-learning methods for identifying plant diseases rely on custom criteria and rarely handle enormous amounts of data. To address this problem, this paper offers an ontology-based method for modelling plant diseases. According to the suggested method, pre-trained architectures include Alex Net and VGG19 CNNs. In order to extract the most useful features from a dataset, the algorithm modifies the details. Based on the correlation coefficient for each characteristic, the right subset of features is selected Convolutional neural networks, artificial neural networks, deep learning, and other artificial intelligence techniques have made it possible to detect diseases in crops like rice, wheat, maize, cotton, tomatoes, peas, potatoes, cucumbers, cassava, berries, peaches, grapes, apples, sweet paper, tea, and others. This study established that training large, publicly available data sets with machine learning algorithms is an effective method for diagnosing plant diseases.