Related Experiment Video
Updated: Aug 10, 2026
11:37
RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
Published on: August 8, 2017
An RGB-D time-series dataset of white button mushroom growth for instance segmentation
1Department of Agricultural and Biological Engineering, Gulf Coast Research and Education Center, University of Florida, Institute of Food and Agricultural Sciences, Wimauma, FL 33598, USA.
Data in Brief
|August 9, 2026
Summary
Accurate mushroom cap detection and segmentation are crucial for robotic harvesting. This dataset aids computer vision research in autonomous mushroom farming by providing RGB-D images and annotations for precise robotic manipulation.
Area of Science:
- Agricultural Robotics
- Computer Vision
- Machine Learning
Background:
- Robotic harvesting of white button mushrooms (Agaricus bisporus) faces challenges in accurately detecting and segmenting individual mushroom caps in dense cultivation beds.
- Precise boundary delineation is essential for size estimation, grasp planning, and collision-free manipulation in automated harvesting systems.
Purpose of the Study:
- To support research on computer vision-based perception systems for autonomous mushroom harvesting.
- To provide a dataset for training and benchmarking algorithms in instance segmentation, growth analysis, and perception for agricultural robotics.
Main Methods:
- A dataset of 129 time-series RGB and depth images of mushrooms was collected from a top-view perspective in an indoor production environment.
- Each image sample includes RGB, depth data, and pixel-level segmentation annotations in COCO JSON format.
- Depth information allows for the extraction of geometric features like cap height, curvature, and inter-mushroom positioning.
Main Results:
- The dataset enables the development of algorithms for accurate instance segmentation of individual mushrooms.
- It facilitates the analysis of mushroom growth patterns and spatial distribution.
- The data supports advancements in perception for agricultural robotics in complex, dense settings.
Conclusions:
- This dataset is a valuable resource for advancing autonomous mushroom harvesting through improved computer vision and perception systems.
- It will aid in the development of more sophisticated agricultural robots capable of precise manipulation in challenging environments.
- The dataset contributes to RGB-D scene understanding in complex agricultural settings.
