Black Pepper Leaf Disease Image Dataset

P. T. Sreethu, Manju Mary Paul, Pratheesh P. Gopinath · 2025 · Image

Description

This repository contains a collection of 1193 original images of black pepper (Piper nigrum L.) leaves collected from different pepper-growing regions of Kerala, India, as part of a research study conducted at Kerala Agricultural University. The dataset comprises four categories: Healthy Leaves, Initial Anthracnose, Advanced Anthracnose, and Phytophthora Foot Rot. Images were captured under diverse field conditions using digital cameras and smartphones, representing variations in lighting, backgrounds, disease symptoms, and environmental conditions. The dataset was developed to support research on automated plant disease identification using machine learning, deep learning, and computer vision techniques. The original images were subsequently augmented to generate 18,234 images for developing and evaluating Convolutional Neural Network (CNN) models. This repository aims to facilitate scientific research, model development, comparative evaluation, and advancements in AI-assisted agricultural disease diagnosis. Researchers using this dataset are encouraged to acknowledge the original publication: Sreethu, P. T., Paul, M. M., Gopinath, P. P., Shahana, I. L., & Radhika, N. S. (2025). Foliar symptom-based disease detection in black pepper using convolutional neural network. Phytopathology Research, 7, Article 21. https://doi.org/10.1186/s42483-024-00305-1.

Methodology

The dataset comprises 1,193 original images of black pepper (*Piper nigrum* L.) leaves collected from different pepper-growing regions of Kerala, India, during summer and rainy seasons. Images were captured using digital cameras and smartphones under varying field conditions, lighting, and backgrounds. With the assistance of plant pathology experts, the images were classified into four categories: Healthy, Initial Anthracnose, Advanced Anthracnose, and Phytophthora Foot Rot. For the associated research, image augmentation techniques, including rotation, flipping, brightness adjustment, and skewing, were applied to generate 18,234 images for CNN-based disease classification. Detailed data collection procedures, preprocessing methods, and experimental protocols are documented in Sreethu et al. (2025), *Phytopathology Research*, 7, Article 21. https://doi.org/10.1186/s42483-024-00305-1.

Crops: Black pepper

Coverage: The image dataset was collected from the Pepper Research Station, Panniyur, Kannur, and black pepper-growing fields across Wayanad, Kasaragod, Kollam, Malappuram, and Thiruvananthapuram districts of Kerala, India. Data collection covered diverse agroecological conditions during summer and rainy seas, All Kerala, Kerala, India

Licence: Custom institutional terms. File downloads require a verified account, contact profile and licence acceptance.

Citation

Sreethu, P. T., Paul, M. M., Gopinath, P. P., Shahana, I. L., & Radhika, N. S. (2025). Foliar symptom-based disease detection in black pepper using convolutional neural network. Phytopathology Research, 7, Article 21. https://doi.org/10.1186/s42483-024-00305-1

Record: DR-wFOz73fHGnSTAgoPuozv