African Elephants, the world’s largest mammal, are under threat across Africa – from poachers and loss of habitat.
To protect elephants, it is important to monitor their numbers and movements.
This is challenging due to the scale of terrain they move over. Current methods vary from the high tech (GPS and aerial photography) to low tech (measuring the amount of dung). However, this data is not enough to monitor the wide range of herds across the continent. A heard can travel over 50km (30 miles) a day.
To help resolve this satellite imagery is now being tested to detect the movement of elephants.
The short version is: Convolutional neural networks can auto-detect elephants in satellite imagery – which then enables tracking, which in turn helps to protect the elephants.
Articles on the techniques are available her
