Historically, medicinal plants have been extensively researched and evaluated because of their critical role in maintaining human health. But identifying medicinal plants is a laborious, time- consuming process that needs to be done by a qualified expert. Thus, a system based on vision can assist researchers and common people in swiftly and precisely identifying herb plants. Therefore, by creating an autonomous convolutional neural network (CNN), this study suggests an intelligent vision-based system to identify herb plants and disease detection. Our results demonstrate the potential of mobile-based solutions to transform on-field agriculture practices by highlighting the efficacy of deep learning in the diagnosis of herb-plant diseases. This research integrates accessibility with cutting-edge technology to support continuing efforts to improve crop health monitoring and reduce output losses resulting from plant diseases. A deep-learning-based technique for detecting plant diseases from leaf images is provided in this research. A diversified image dataset of plant leaves with 10 distinct plant leaf in 4236 images was used for herbs plant work