An Artificial Intelligence Ready Dataset for Asian Citrus Psyllid Detection

The dataset contains high-resolution images of yellow sticky traps used for the collection of Asian Citrus Psyllid (ACP) in Mediterranean agro-ecosystems. The traps were placed under both controlled and field conditions to capture realistic variability in insect posture and density, while retaining representations of non-target insects and natural debris. Each image is accompanied by manually curated annotations, including bounding boxes and pixel-level segmentation masks, enabling the use of the dataset across multiple computer vision tasks such as object detection, instance segmentation, and small-object recognition. Additionally, the dataset contains systematically generated augmented variants, with explicit traceability preserved between original images and derived counterparts. The aim of this work is to support the development and evaluation of artificial intelligence (AI) models for the detection and monitoring of the Asian Citrus Psyllid (ACP). Dataset Composition The dataset contains: 200 high-resolution images (2198 × 1440 pixels) 27,221 annotated objects, including: 3,062 ACP instances - Class 0 24,159 non-target objects (other insects and debris) - Class 1