Deep learning applied to agriculture

Cases

The climate is changing, there is less agricultural land, and consumer demands are shifting. All of this calls for innovation and greater efficiency in agriculture. The mushroom sector is at the forefront in this area, as can be seen at Verbruggen Paddestoelen. They recently carried out an exploratory study into the possibilities of machine learning for sorting and packing products in collaboration with Arnhem engineering firm QING.

Verbruggen Paddestoelen uses residual flows from agriculture as raw materials, and cultivation takes place in climate cells. This so-called vertical farming makes efficient use of the available space and creates more control over the conditions, so less land and water is needed. 

Popular as meat substitute

The company from Erp is currently focused on oyster mushrooms, which are becoming increasingly popular as a meat substitute due to their firm and meaty structure. Harvesting, packing and sorting oyster mushrooms still involves a lot of manual labour and finding suitable personnel keeps becoming more difficult. 
The company is looking for automation opportunities to meet the increasing demand, starting with the sorting and packing process. According to the high-tech engineering firm on its website, the oyster mushroom grower has entered into a collaboration with QING to this end. Together, they investigated whether it is possible to analyse the oyster mushrooms with image processing reliably. 

Automated packing machine

This research provided the two parties with important insights into applying conventional image processing and deep learning to extract more value from the mushrooms. The study also provided insight into how significant the investment for an automatic packing machine would be. Verbruggen still has to decide how they are going to roll out this innovation. 

Want to know more about AI applications in the agriculture and food industry? Contact us.

AI hub East Netherlands is partner of