The RGB camera method is hugely economical compared to traditional multispectral cameras.

A recent study from the University of Illinois and Mississippi State University (MSU) suggests that artificial intelligence (AI) combined with standard red-green-blue, or RGB cameras can provide the equivalent crop prediction tools as expensive multispectral cameras. This procedure can dramatically reduce costs of crop surveillance.

The traditional multispectral cameras provide color maps representing vegetation to monitor plant health and spot problems by displaying healthy areas as green, and problem areas as red. In contrast, RGB cameras lack the ability to capture near-infrared light that is critical for plant health assessment. The researchers used a neural network called Pix2Pix to translate RGB images into NDVI and NDRE color maps with red and green areas by training the network on a large number of multispectral and regular pictures. The AI network can extract information from the green channel by training on the near-infrared channel, making it possible to generate the required vegetation indices using only the RGB camera, mounted on a low-cost drone, a combination which significantly reduces the cost. (PHYS.ORG)

Crop specialists were asked to compare AI-generated and multispectral images, and they found no observable differences, indicating similar predictions from both.

The aerial imagery obtained from drones provides valuable information on crop health for management decisions, including spotting areas of storm damage or where additional inputs are required of nutrients.

The AI software and procedures used in the study are available for organizations that want to avail of its benefits. They may also utilize the software to train their network on additional datasets. The study highlights the potential of AI in reducing costs, making it accessible to everyone, and providing farmers with valuable information to boost yield and take better care of their land.

Related Story: LG’s IT Group To Set Up AI Platform For Outdoor Smart Farming

Image Credit: Flickr