New AI Tool Can Help Track And Protect Wildlife, Has Already Mapped 4.5 Million Wild Animals
Google, along with seven organisations led by Conservation International have mapped over 4.5 million animals in the wild by capturing their images on motion-activated cameras. These images are then fed into an AI-enabled Google cloud-based platform that is helping simplify the monitoring of wildlife.
Artificial Intelligence is helping people make their lives better in various fields like medical sciences, helping us combat climate change, securing our lives on the cloud, and now it is also being helped to save the wildlife.
Google, along with seven organisations led by Conservation International have mapped over 4.5 million animals in the wild by capturing their images on motion-activated cameras. These images are then fed into an AI-enabled Google cloud-based platform that is helping simplify the monitoring of wildlife.
The AI system is one of the most diverse in the world and allows people to look at millions of camera-trap images and filters it as per species, country and year. A normal human would taken an hour to look through 300 to 1000 images. However, the AI does the same 3000 times faster at a rate of 3.6 million photos per hour.
The AI is trained to automatically classify species in an image using Google¡¯s open-source TensorFlow framework.
It is understandable that identifying animals on the basis of species can get challenging. The AI model is trained with over 614 species like jaguars, while-lipped peccaries and African elephants. The AI has a success rate of 80 to 98.6 percent.
Google said in a statement, "With photos and aggregated data available for the world to see, people can change the way protected areas are managed, empower local communities in conservation and bring the best data closer to conservationists and decision-makers."
With this system, people working for the protection of specific species can keep a track on their number on the click of a button -- filtering location, species, dates etc to get a more specific data.