Modern agriculture and horticulture are often labor intensive. At the same time, while wages are on the rise, the availability of a skilled workforce that is willing to perform the repetitive tasks in the harsh environmental conditions of a greenhouse, orchard or a field, is decreasing rapidly. Interestingly enough, however, the current state of the art in automated harvesting of fruits and vegetables has remained remarkably stationary in the past decades, in part because of technological barriers that have not been addressed successfully thus far. Contributing to the effort to revolutionize agriculture, our group has been involved in various research projects in agrobotics, focusing on the where agriculture and computer vision meet. What we term agrovision offers challenges that no other domain in computer vision can offers, with tangible potential to aid society as a whole.
While smaller projects preceded it, our first major contribution was in the EU-FP7-project cRops , where extensive research has been performed by the 14 partners on generic robotic solutions for automatic selective harvesting. One of the main applications of cRops was for sweet pepper harvesting, which led to the new EU project SWEEPER – a Sweet Pepper Harvesting Robot. Our goal is to develop a fully autonomous robot that can harvest sweeper peppers in greenhouses 24/7. The project incorporates 6 partners from 4 different countries (The Netherlands, Belgium, Sweden and Israel) and involves experts from a wide-range of disciplines, including horticulture, horticultural engineering, machine vision, sensing, robotics, control, intelligent systems, software architecture, system integration and greenhouse crop management.
Polina Kurtser & Yael Edan (2018). “Statistical models for fruit detectability: spatial and temporal analyses of sweet peppers”. Biosystems Engineering, pp. 272–289