Pepper and maize, vital ingredients in many dishes, are continuously at risk from diseases that can affect crop yield and food security. AI is essential to 21st-century life. It is used for medical applications, communication, object detection, identification, agricultural and tracking. This paper uses deep learning to diagnose bell pepper plant diseases in large fields. Most bell pepper producers are unaware of bacterial spot disease. Disease propagation often reduces harvest output. Early detection of bacterial spot disease in bell pepper plants is essential for treatment. Quickly and effectively diagnosing pepper and corn crop diseases prevents productivity losses and optimizes crop management. This study examined how well CNNs and YOLOv7 classified common agricultural diseases. The suggested method includes image pre-processing, feature extraction, and classification. A large collection of pepper and corn images showing visual features and illnesses was assembled. Both models were trained and tested for illness classification. We used a large dataset of pepper and corn pictures modified with disease severity and backdrops. CNNs classified peppers at 99.4% and corn at 98.8%. Despite having poorer classification accuracy (97.6% for peppers and 99.92% for maize), YOLOv7 can detect unhealthy areas in real time and generate bounding boxes. This allows precise pesticide or fungicide application, decreasing environmental impact and optimizing efficacy. Both methods show promise for pepper and corn disease prevention, according to our research.