Machine learning is constantly unleashing tremendous potential in an immense number of applications, notably the internet of things (IoT), computer vision, natural language processing, and many more which are used to ensure higher accuracy of computer vision in the agriculture industry. A detecting of plant disease detection system is a comprehensive framework that leverages computer vision and machine learning techniques to identify plant diseases from images. Infectious agents find their natural reservoirs in soil, water, and animals, with insects playing a significant role. Pathogenic organisms, including fungi, bacteria, viruses, protozoa, insects, and parasitic plants, primarily contribute to the occurrence of infectious plant diseases. The architecture incorporates various critical aspects, which might involve gathering information, preprocessing and the process of augmentation, feature extraction, model training, evaluation, and validation. Convolutional neural networks (CNNs) are commonly used for model development. This research proposes a novel approach, which has been successfully utilized to tackle several optimization problems, to find out the conceptual framework of a Bayesian network utilizing a discrete firefly optimization technique.