Animal detector demo
Image recognition can be used to automatically detect a wide range of animals from alpacas to zebras. With the use of artificial intelligence it is possible to identify, count or follow specific animals. To automate these tasks it will save scientists time. It can even help with safety issues in for example zoos or protected areas. To show all of the possibilities, we have made an online demo where you can test your own image and see how the standard algorithm works for different kinds of species.
Common Objects in COntext
In the example above, the image is trained with the COCO dataset and it will calculate the possibility that the object is an animal. One example of the projects of Studio diip that containted animals and image recognition was called Fish on Wheels: an electrical car that was driven by a fish. The learning algorithm that we use here is trained to detect the following species:
- Birds
- Cats
- Dogs
- Horses
- Sheep
- Cows
- Elephants
- Bears
- Zebras
- Giraffes
Common Objects in COntext
In the example above, the image is trained with the COCO dataset and it will calculate the possibility that the object is an animal. One example of the projects of Studio diip that containted animals and image recognition was called Fish on Wheels: an electrical car that was driven by a fish. The learning algorithm that we use here is trained to detect the following species:
- Birds
- Cats
- Dogs
- Horses
- Sheep
- Cows
- Elephants
- Bears
- Zebras
- Giraffes
Image recognition with deep learning
For this example we have used an algorithm that is trained on a thousand photos with all sorts of animals from the COCO dataset (Common Objects in COntext). This database includes images of the chosen animals where you can exactly see where the mammal is hidden in the photo. Therefore, a learning algorithm can train itself to automatically recognize different spiecies in different photos. Off course you can also use this technique to detect other objects as well. We use tools like Keras and TensorFlow to see how well the algorithm performs.