Sclerotinia and White Mold (SWM) dataset

Description

The Sclerotinia sclerotiorum fungus and white mold infect common bean (Phaseolus vulgaris L.) crops and over 400 hosts, including soybeans, cotton, sunflower, canola, and tobacco. These crops have high value and significant economic and social impacts across countries because the fungus persists in the soil for long periods, and the disease is epidemic and easily widespread. The SWM dataset addresses a gap in this disease by comprising high-resolution images from common bean crops. The images were captured in real field conditions, providing a diverse set of images with varying lighting, angles, and backgrounds. The dataset was carefully annotated by Phytopathologists, providing annotations in YOLO and Pascal VOC formats. This dataset was specifically designed for training, validating, and testing object detector models to support research in plant disease detection and precision agriculture applications. The annotations refer to three lifecycle stages: mature sclerotium, apothecium, and white mold. The class labels for the lifecycle stages are 3, 1, and 5, respectively. The image set is organized into two datasets: the SWM dataset and the R-SWM dataset. The SWM dataset contains 2,527 images in JPG format with dimensions of 300x300 pixels, balanced across the three classes, and split into training, validation, and test sets. The R-SWM dataset contains 1,278 original images in JPG format with variable dimensions, along with annotations for the three classes, split into training, validation, and test sets.

Institutions

• Universidade Estadual de Campinas (UNICAMP)

  São Paulo, Campinas
• Universidade Federal de Goiás

  Goiás, Goiânia
• Wageningen University & Research

  Gelderland, Wageningen
• Brazilian Agricultural Research Corporation

  Federal District, Brasília

Categories

Licence