Deep learning to deny poaching - SecurityBis
Deep learning

Poaching is an epidemic in southern Africa. To watch many beautiful animals, particularly rhinos and elephants, decline in numbers at a rapid rate is a tragedy.

There are various poaching detection technologies available including perimeter, ground, aerial, sensor and animal tagging. They all have their strengths but poaching statistics keep increasing and the need for new approaches that provide better protection against poaching is urgent.

Madikwe Game Reserve is currently the fifth largest game reserve in South Africa and has to work hard to protect its wildlife from those who want to poach and harm the animals, as well as to protect those who work to keep them safe.

Security BIS is a distributor of business solutions in South Africa and the SADC region. AxxonSoft is Security BIS’ flagship security camera monitoring and management system that uses cutting-edge analytics technology to give clients the edge in crime and incident prevention. Security BIS offers resellers a full consultation, design and implementation strategy; employing certified AxxonSoft technical engineers.

Security BIS was tasked with enhancing the security and surveillance operations at Madikwe to prevent poaching.

Some perimeter security systems are combined with video surveillance, but they often prove ineffective. Since animals often bump the fence, such systems generate false alarms, so security staff can’t possibly monitor all the events and easily miss real threats when poachers invaded the park. This is not the case with AxxonSoft.

Proactive surveillance with AI

AxxonSoft’s deep learning-based artificial intelligence technology provides proactive surveillance in identifying poaching threats across southern Africa to help combat and prevent poaching by distinguishing between animals and humans and picking up on false alarms.

Over 100 cameras now track the movement of animals, people and cars in Madikwe, using facial and licence plate recognition technology and object tracking. The system uses deep learning technology to detect the difference between humans and animals.

As soon as there is a breach, the surveillance control centre is immediately notified. Cameras are used to identify where the breach has occurred. The control centre immediately gets all the necessary information and the anti-poaching unit is dispatched. This has proved very successful in preventing killings as the team is able to get to the scene of the crime quickly.

International assistance

AxxonSoft’s global team worked tirelessly to create a system that addresses poaching head on. They pioneered a new video surveillance implementation of deep learning technology that could detect the difference between humans and animals.

By installing cameras around the perimeter of parks and game reserves, security forces can adapt their approach to poaching. Rather than constantly trying to apprehend criminals in the aftermath of poaching incidents, they can immediately send rangers and/or camera drones to hotspots when alerted by a smart software system that red flags likely poachers before a crime is even committed. Rangers can move from a reactive to a proactive response.

Thanks to AxxonSoft’s solution based on self-learning and neural networks, Madikwe can now differentiate between animals and people to prevent poaching. This anti-poaching solution aids in drastically reducing the upsetting poaching statistics.