Tomorrow’s agricultural network depends on devices and automation. The camera is an important element, and artificial intelligence is a central technology here. Intelligent applications such as harvesting robots can contribute significantly to this.
Lettuce is a valuable crop in Europe and the United States. But labor shortages make it difficult to procure vegetables from this important field, as one of the major challenges in the sector is finding sufficient seasonal labor to meet harvest promises. Moreover, since wage inflation is rising faster than producer prices, margins are very narrow. In England, engineering and agricultural machinery experts are developing a robotic solution for lettuce crop automation in collaboration with IDS Imaging Development System GmbH.
The team is working on a project funded by Innovate UK and includes experts from Landmaschinenfabric Grimm, the Agri-EPI Center (Edinburgh, UK), Harper Adams University (Newport, UK), and the Center for Machine Vision at the University of the West of England. Bristol) and the UK’s two largest salad producers, G’s Fresh and PDM Productions.
This project makes it possible for existing machines responsible for cutting leaks to lift lettuce from the ground and pinch between compression bands. The outer leaves of lettuce are mechanically removed to reveal the stem. Using industrial image processing and artificial intelligence, a precise cutting point on the stalk is then determined to separate the lettuce head from the stem.
According to IDS Product Sales Specialist Rob Webb, “Cutting an iceberg lettuce is the most technically complex step in the automation process, according to team members of subsidiary Salad Harvesting Services Limited.” A GigE Vision camera from the uEye FA family has been integrated into the prototype of the harvesting robot. It is considered to be particularly resistant and therefore ideal for the needs of the environment. “Since this is an outdoor application, an IP65 / 67 enclosure is required,” Webb points out.
Tackling the challenges of tomorrow’s agriculture
The choice falls on a uEye FA model, equipped with Sony’s compact 2/3 ″ CMOS sensor Global Shutter IMX264. “The sensor was initially chosen for its versatility. We don’t need full resolution for AI processing, so sensitivity can be increased through compartmentalization. The larger sensor layout also eliminates the need for wide-angle optics,” Rob Webb said. In use, the CMOS sensor captivates with its exceptional image quality, light sensitivity and exceptionally high dynamic range. It delivers 5 megapixel images at a ratio of 5: 4 at 22 frames per second, virtually sound-free and with very high contrast, even in applications with unstable lighting conditions. The energy chain compatible lens tubes and wide accessories with cables are as strong as the camera housing and screw connectors (8-pin M12 connector with X-coding and 8-pin binder connector). Another advantage: internal camera functions such as pixel pre-processing, LUT or gamma reduce the required computing power to a minimum.
“We are pleased with this participation in the project and we look forward to the results. Jan Hartmann, managing director of IDS Imaging Development Systems GmbH, said: “We are confident of the potential for automation and efficiency enhancement of lettuce collection, not just to meet seasonal labor shortages.
The challenges in the agricultural sector are really complex. According to the Food and Agriculture Organization of the United Nations (FAO), due to the dramatic increase in population, agricultural productivity will increase by about 50% by 2050 compared to 2012. The prospect of such a return represents a huge challenge for the agricultural sector, which is still at the very beginning in terms of digitization compared to other industries and which is already under tremendous pressure to innovate in the face of change. Climate change and labor shortages. Tomorrow’s agricultural network depends on devices and automation. The camera is an important element, and artificial intelligence is a central technology here. Intelligent applications such as harvesting robots can contribute significantly to this.