UAE smart feeder processes 15 million images to monitor wildlife feeding

A UAE-developed smart wildlife feeder is using artificial intelligence to identify animals, monitor their numbers and consumption, and automatically adjust feed quantities based on real-time data.

Developed by Dub Dev Technology, the "Mghzlan Smart Feeder” combines computer vision, Internet of Things technology and solar power, allowing wildlife feeding and monitoring operations to be managed remotely.

The system processes more than 15 million images annually, with field monitoring recording animal classification confidence levels of between 80 and 97 per cent in most cases. It has dispensed more than 50 tonnes of feed in a year and has been deployed in more than three countries.

When animals approach the feeder, day-and-night cameras identify their species and numbers and record their arrival times.

The system also monitors feed consumption and remaining stock, providing data that enables quantities to be tailored to the needs of each location.

Ali Khalfan Al Gaz Al Falasi, Director of Dub Dev Technology, told the Emirates News Agency (WAM) that the device automatically dispenses calculated quantities of feed at scheduled times, reducing the need for daily human presence and limiting unnecessary disturbance to wildlife.

He said the AI and computer vision system, including species-recognition algorithms, the data platform and control dashboard, was developed entirely in the UAE.

Field tests achieved confidence levels of 80 % - 97 % in identifying species including gazelles, oryx, antelopes and birds, even in crowded or low-light conditions, he added.

The platform also tracks animal activity patterns, feed consumption, stock levels, temperature and humidity.

Operators can remotely manage multiple units and receive alerts when feeders require replenishment or when people or vehicles are detected near operating sites.

The technology has been deployed at Al Marmoom Desert Conservation Reserve in Dubai. The company has also developed a version for feeding birds, while its computer vision technology can identify other wildlife, including Arabian oryx, antelopes and rabbits.

By integrating wildlife, feed and environmental data on a single platform, the system aims to improve resource efficiency and support more sustainable management of wildlife habitats while reducing unnecessary human intervention.

WAM