# An annotated image dataset for small apple fruitlet detection in complex orchard environments

## TL;DR

This paper introduces a dataset of annotated images for detecting small apples in orchards, aiming to improve automated thinning systems.

## Contribution

The novel contribution is a standardized, annotated image dataset capturing real-world orchard conditions for small apple detection.

## Key findings

- The dataset includes 2,517 RGB images with annotations in PASCAL VOC and YOLO formats.
- Validation experiments across multiple detection frameworks confirmed the dataset's effectiveness.
- The dataset supports automation in apple thinning and improves fruit quality.

## Abstract

This study introduces a small apple pre-thinning dataset designed to support the development of intelligent thinning systems by providing reliable data for small apple detection. The dataset comprises 2,517 RGB images (original size 3024×3024 pixels, uniformly resized to 500×500 pixels for standardization) systematically captured under real-world orchard conditions. The dataset encompasses natural variations in weather conditions (sunny/cloudy), lighting scenarios (direct sunlight/backlight), and fruit sizes (3-25mm diameter range) to ensure broad applicability. Each image was meticulously annotated using LabelImg software, with all small apple targets precisely labeled using both PASCAL VOC (XML) and YOLO (TXT) format bounding boxes, facilitating compatibility with various detection frameworks. Validation experiments conducted across multiple detection architectures (including Faster R-CNN, Cascade R-CNN, YOLO series, RT-DETR, DEIMv2, etc.) demonstrate the dataset’s effectiveness. This dataset serves as a valuable resource for developing intelligent thinning systems, with potential applications in promoting automation in the apple industry, enhancing thinning efficiency, and improving fruit quality.