AI for habitat recognition
High image quality and algorithms facilitate the work
Automated nature recognition is an example of one of the major digitization projects that has got off to a good start in a short time using the JUMP Digital Factory methodology.
National legislation and EU directives require the Danish Environmental Protection Agency to monitor and protect various habitat types, and one of the tasks for the Danish Environmental Protection Agency's biologists is, among other things, to look after nature by the Wadden Sea. Every six years, biologists have studied the versatile nature of the Wadden Sea, but it is both difficult and costly to plan for the tide, which exposes and floods the seabed twice daily.
“We have tested drones for a short period, but quickly found out that using both existing aerial photos and data from the Danish Agency for Data Supply and Efficiency, both in terms of time and finances. We have also developed an advanced image recognition algorithm that can quickly help us show and delineate the habitat types in a given area. Something that today is done manually, ”says Lars Møller Christiansen, deputy director at the Danish EPA.
"The project is showing a lot of promise, and the initial results have been so convincing that we will now also test the method for monitoring other habitat types across the country," he says.