In a breakthrough study, researchers at Penn State have developed a machine vision system capable of locating and identifying the key “king flowers” within clusters of apple blossoms on trees in orchards. This is a critical step in the development of a robotic pollination system that could revolutionize apple production.
Traditionally, apple productivity has relied on insect pollination, but with honeybee populations around the world declining at alarming rates due to colony collapse disorder, producers are in need of alternative methods of pollination. The study, led by Long He, assistant professor of agricultural and biological engineering, is the latest in a series of projects by He’s research group to develop robotic systems for labor-intensive agricultural tasks such as mushroom picking, apple tree pruning, and green-fruit thinning.
The primary goal of this project was to develop a deep learning-based vision system that could precisely identify and locate king flowers in tree canopies. Xinyang Mu, doctoral student in the Department of Agricultural Biological Engineering, used Mask R-CNN, a popular deep-learning computer program that performs pixel-level segmentation to detect objects that are partially obscured by other objects, to identify and locate the king flowers.
To build the detection model, Mu captured hundreds of apple blossom cluster photos and developed a king flower segmentation algorithm to identify and locate the king flowers from the raw dataset of apple flower images. The research was conducted at Penn State’s Fruit Research and Extension Center, Biglerville, using Gala and Honeycrisp apple varieties.
Training the machine vision system to locate king flowers was challenging, as they are the same size, color, and shape as the lateral blossoms in clusters, and the king flowers are typically obscured by surrounding flowers because of their central position. However, the researchers reported a high level of king flower-detection accuracy, with results varying from 98.7% to 65.6% compared to measurements taken manually by researchers identifying king flowers by eye.
“We think this result will provide baseline information for a robotic pollination system, which would lead to efficient and reproducible pollination of apples to maximize the yield of high-quality fruits,” He said. “In Pennsylvania, we still can rely on bees to pollinate apple crops, but in other regions where bee die-offs have been more severe, growers may need this technology sooner than later.”
This revolutionary machine vision system has the potential to greatly improve apple production and make it more sustainable and efficient, helping to ensure a steady supply of high-quality fruits for consumers. It is a major step forward in the development of robotic pollination technology and a promising solution to the challenges facing apple growers.