Significant progress in bird image capture with AI in the IA4Birds project

The IA4Birds project is an AI-based platform that uses audiovisual devices to monitor various bird populations. This allows for the prevention of threats and risks they face, as well as the assessment of whether a location is suitable for the development of a wind farm. The goal is to help reduce the high mortality rate of birds, such as those caused by collisions with wind turbines.

Coordinated by AIR Institute and funded by Fundación Biodiversidad, the initiative also involves collecting various information sources and creating a database.

To achieve its objectives, the project has been using the AXIS Q6225-LE PTZ camera for several months, a high-quality device with features that allow for accurate bird detection in the area of interest. Its HDTV 1080p resolution, ½” sensor, and 31x optical zoom enable detailed capture from long distances.

The AI team behind the initiative has made progress with the imaging module, where two studies have been carried out simultaneously: one on distance and the other on zoom. The first study estimates the distance at which birds can be found, considering their actual size, the applied zoom, and their species-specific pixel size. The zoom study, conducted using the YOLOv8 tool, delves into the optimal zoom to be used, based on the number of birds detected and their size. Below are the captured images of storks and red kites.

The IA4Birds project is an initiative supported by AIR Institute and Fundación Biodiversidad from the Ministry for Ecological Transition and the Demographic Challenge (MITECO), under the framework of the Recovery, Transformation, and Resilience Plan (PRTR) funded by the European Union - Next Generation EU.