Capsicum (bell pepper) is a globally important crop whose productivity is severely limited by leaf diseases, including bacterial spot, anthracnose, mosaic virus and powdery mildew. Early and accurate detection of these diseases through non-destructive imaging and machine intelligence is critical for yield protection. Recent advances in deep learning, particularly convolutional neural networks (CNNs), have revolutionized plant disease identification, enabling automated pipelines that analyse leaf images and classify disease. This survey focuses specifically on the detection of Capsicum leaf disease using deep learning. We review major Capsicum-related image datasets (e.g. PlantVillage pepper images, the COLD chili-onion dataset, the BellCrop dataset and other recent curated collections) and summarize state-of-the-art deep models applied to Capsicum disease classification. Original figures illustrate a typical detection pipeline and a CNN architecture. We also compare model performances reported in the literature. A rich literature review (65+ open-access, recent references) highlights CNN-based classifiers, transfer learning approaches and case studies in Capsicum disease detection. This work serves as a detailed reference for researchers and practitioners developing AI systems for the early detection of pepper disease.
