In today’s agriculture, leaf disease is a major issue. It hinders the natural growth of plants. Stifles a country’s economic development. It will lower the quality of agricultural products. Leaf disease can develop as a result of bacterial, fungal, or other causes. Finding and detecting sick plants in the open eye takes a long time. As a result, automatically detecting and resolving plant disease is critical. Pepper Bacterial spot disease is generally caused by Xanthomonas campestris which reduces pepper production and quality. In this paper, we used the plant village dataset. The dataset contains 2080 image data of bacterial spotted pepper bell leaf and 1,881 healthy bell pepper leaf. We will study pepper belt bacterial spot disease. Using Conventional Neural Network(CNN), our suggested method will detect bell pepper bacterial spot plant disease. We’ll classify each plant image and look for illnesses. Our proposed CNN system has an accuracy rate of testing is 96.88% the accuracy rate of training and validation is 99.44% and 97.34% respectively.