Sign up for our daily Newsletter and stay up to date with all the latest news!

Subscribe I am already a subscriber

You are using software which is blocking our advertisements (adblocker).

As we provide the news for free, we are relying on revenues from our banners. So please disable your adblocker and reload the page to continue using this site.
Thanks!

Click here for a guide on disabling your adblocker.

Sign up for our daily Newsletter and stay up to date with all the latest news!

Subscribe I am already a subscriber

Researchers develop an AI system that allows greenhouses to operate without human intervention

This technology is a step forward for future space missions
One of the challenges to colonize other planets and establish human settlements, starting with the planets closest to Earth in our Solar System, such as Mars, is to guarantee food for their future inhabitants. Since these planets have different atmospheric conditions than Earth, the fresh produce would have to be grown in greenhouses with a controlled atmosphere. The development of technologies based on artificial intelligence and computer vision for the automation of agricultural tasks in these types of greenhouses is one of the priority research objectives for future missions on Mars.

In fact, there is a place where advanced life support systems geared towards the maintenance of greenhouses outside Earth are already being tested. It is an autonomous plant cultivation module that is operating at the German base Neumayer III, located in Atka Bay, Antarctica.

Scientists from the Skolkovo Institute of Science and Technology (Skoltech) in Russia and their collaborators at the German Aerospace Center (DLR) have developed an artificial intelligence system that allows greenhouses to operate autonomously without human intervention, through image processing of plants in greenhouses, monitoring of vegetable growth, and automation of agricultural tasks.

Sergey Nesteruk's team processed a collection of images from remote automated systems using their new approach based on convolutional neural networks and outperforming the most popular codecs in reducing image size without apparent degradation of the image quality.

The researchers used the information from the reconstructed images to train a computer vision algorithm that is capable of classifying 18 varieties of plants by species and at different stages of development, with an accuracy of 92.6%.

Nesteruk and his colleagues detail their new system in the article, "Image Compression and Plants Classification Using Machine Learning in Controlled-Environment Agriculture: Antarctic Station Use Case", in the academic journal IEEE Sensors .

 

Source: NCYT de Amazings 

Publication date: