High-Resolution Images of Lychee Plant Diseases for Classification and Detection
Description
This dataset consists of high-resolution images of lychee plants (Litchi chinensis) collected from Biral, Dinajpur, Bangladesh. Lychee is a tropical fruit with significant economic value, particularly in Asia, known for its sweet and juicy flesh as well as its rich nutritional content. Monitoring the health of lychee plants and identifying disease stages is crucial for optimizing agricultural yields, enhancing crop management, and advancing precision farming practices. The primary goal of this dataset is to support the development of machine learning and deep learning-based models for the classification, detection, and monitoring of various disease stages in lychee plants. Unlike datasets focused solely on growth stages, this collection emphasizes the identification of multiple disease conditions affecting lychee plants under natural growth environments. This dataset has undergone a significant update in version 2, where the background of the images has been removed to improve the accuracy and efficiency of machine learning models during training. By isolating the plants and their diseases from distracting background elements, the new version enhances the focus on the plant features crucial for disease detection. The dataset is divided into six distinct disease categories: Anthracnose Cloudy: 971 images Algal Spot Indirect: 697 images Dry Leaves: 477 images Entomosporium Spot: 407 images Leaf Mites Direct: 567 images Mayetiola PostRain: 647 images Total Images: 3,766 images After Augmentation: Anthracnose Cloudy: 1,000 images Algal Spot Indirect: 1,000 images Dry Leaves: 1,000 images Entomosporium Spot: 1,000 images Leaf Mites Direct: 1,000 images Mayetiola PostRain: 1,000 images Total Images: 6,000 images Image Specifications: Original Resolution: 3024 × 4032 pixels Compressed Resolution: 560 × 420 pixels Image Format: JPG This dataset is invaluable for research in agricultural disease management, precision farming, and plant pathology, offering a valuable resource for machine learning and deep learning model training aimed at improving crop health monitoring.
Files
Institutions
- Daffodil International University