Monitoring animal populations is labor intensive, and many researchers are turning to technology to reduce the burden on this important aspect of conservation. Ed Mill, Principal Engineer at Arm, leads a team of researchers aimed at monitoring and protecting brown bears. They have developed an AI-based system that can track these beautiful beasts. Bear ID have plans to consolidate raspberry pith to their setup.
This project specializes in installing remote cameras in areas frequented by these roaming giants. The system they set up uses machine learning to identify bear faces in real time, so they know for sure when a bear has been detected. This will help researchers monitor bears safely from a distance without interfering with their natural migrations.
According to BearID, AI face algorithms detect bears that use their eyes and nose to build a triangle. You can also recognize them by considering the ears. Brown bears have several traits that change with the seasons, such as fur and weight, so it is helpful to focus on these constant traits.
The team recently shared details raspberry pi The blog post explains that most of BearID operates as a Microsoft Azure-based cloud service. In the field, we rely on Arm-based platforms. The team hopes to implement a Raspberry Pi 4 Model B here in the future.
The BearID team shares the code with anyone who wants to check it out. Github. There you’ll find all the details about how it works and the tools used. One of the main systems they rely on is Dlib. Dlib is a C++ toolkit packed with useful machine learning algorithms to help identify and track bears. You can see that much of the code is Python based.
If you want to know more about this project, please visit the official website bearresearch.org Here you can get a behind-the-scenes look at how bears are in action, as well as some cool shots of bears in the wild. Also, if you’re looking for more inspiring microelectronics creations, I encourage you to peruse our list of Best Raspberry Pi Projects.
