Researchers have developed AI that can identify, count and describe wild animals that are printed in photographs captured by motion sensor cameras. The technology that scans these photographs for this information is Deep Neural Networks. Deep neural networks are a form of computational intelligence that requires an enormously huge amount of training data that will help duplicate an animal’s brain – the way they see and understand the world.
Jeff Clune, an Associate Professor at the University of Wyoming in the US says, “This “intelligent” move can help improve our ability to both study and conserve wildlife, and save their precious ecosystems.” The study till now has harnessed 3.2 million labeled images produced by more than 50,000 human volunteers over several years. The labeled images are as a result of manually identifying and naming each image.
The study seems to extend for about a decade more i.e. eight-ten years of extra human-effort to help discover millions of other valuable images that can be redeployed to help other projects. The idea behind this approach of imbibing AI IT service to fight animal extinct is simple, that is when these photographs are converted to text and numbers, it serves as valuable information that can be studied and made useful for several other eco-friendly projects.