SugarcaneLD-BD: A Sugarcane Leaf Diseases Dataset for Classification of Sugarcane Leaf Diseases Using Machine Learning

Description

Sugarcane is susceptible to various leaf diseases that can negatively affect plant health and reduce yield. Early detection and proper management of these diseases are crucial for maintaining crop productivity. SugarcaneLD-BD dataset was compiled to support research on low-cost, non-invasive, and rapid identification of sugarcane leaf diseases using machine learning. It comprises 638 JPG images across five classes: four major sugarcane leaf diseases (a) Red Rot, (b) Red Leaf Spot, (c) Ring Spot, and (d) Eye Spot, and one class representing healthy leaves. Images were collected from four locations in Bangladesh: (a) Bangladesh Sugarcrop Research Institute (BSRI) research field in Pabna (b) BSRI Gazipur Regional Station research field in Gazipur (c) Farmers' field in Narsingdi (d) Farmers' field in Natore Data collection took place between September and October 2023 using two smartphone cameras. All images were labeled and verified by an expert plant pathologist and resized to 224 × 224 pixels. Researchers and practitioners can use this dataset to develop and evaluate machine learning models for sugarcane leaf disease detection.

Institutions

• Bangabandhu Sheikh Mujibur Rahman Agricultural University

  Gazipur, Salna
• University of Dhaka Faculty of Engineering and Technology

  Dhaka District, Dhaka
• Bangladesh Sugarcrop Research Institute

  Rajshahi, Ishwardi

Categories

Funders

• Ministry of Science and Technology

  Dhaka Division, Bangladesh

Licence