Durian Plantation Pink Disease Recognition Image Dataset

In the durian plantation industry, pink disease and other diseases often lead to reduced yield and significant economic losses due to the lack of effective disease recognition methods. Existing disease identification methods largely rely on human expertise, lacking efficiency and accuracy. This dataset aims to enhance the precision and efficiency of disease detection using image recognition technology, meeting the need for rapid disease identification and management by growers. Data is captured in a natural planting environment using high-resolution cameras, ensuring the authenticity and diversity of samples. Data is annotated through multiple rounds by experts to ensure labeling accuracy and consistency; it also undergoes professional review by experts. The annotation team consists of more than 20 experts in plant pathology and data scientists. During data preprocessing, color correction and denoising are performed, and the data is ultimately stored in JPG format for convenient model training.