Dataset for:Enhanced YOLOv8-ECCI Algorithm for High-Precision Detection of Purple Spot Disease in Soybeans

Soybean disease image data were collected at the Yangjiang Customs, China, on April 3, 2024, between 15:00 and 17:00 (UTC+8). All images were captured in a laboratory setting with a fixed camera position at a resolution of 3024 × 4032 pixels. To simulate real-world conditions and enhance data diversity as well as model generalization capability, the soybean samples were systematically repositioned for each capture. The primary symptom of the soybean disease, identified and confirmed by plant pathology experts, is purple blotching. This dataset was subsequently augmented to a total of 900 images using data augmentation techniques. The complete dataset was then randomly partitioned into training, validation, and test sets at a ratio of 7:2:1. The training set was utilized for model training, the validation set for preliminary model performance evaluation and hyperparameter tuning, and the test set for assessing the final model's classification accuracy and generalization capability.