Researchers of the Jena University in Germany have developed an algorithm that allows the software to memorize the appearance of a variety of animal and plant species and thereby developing the ability to discriminate between them. The visual fine-grained project not only aims to develop a new precise way of pattern recognition but also to deliver the parameter the software develops during the learning process.
The algorithm uses deep learning and big data methods for pattern recognition and by including images of many different perspectives it can recognize e.g. birds and dogs even while they move.
The algorithm we developed is able to automatically distinguish between different types of birds, flowers, dog breeds, and in general very similar object categories. It is based on computer vision and machine learning techniques that learn the appearance of object categories from a given set of images together with their annotations. Very recent ideas from the deep learning area allow for estimating very complex visual models and boost the recognition performance up to 82% for a dataset with 200 different bird categories.
One potential outcome would be an app that would enable all of us to identify species in the field as using only a cell phone camera. There is already one less versatile one available. The Merlin app, a tool used in North American to identify birds now lets you upload an image of a bird that you’ve photographed, and if the photo shows one of the supported species, it returns the correct species in the top 3 results, 90% of the time. It currently supports 400 species in North America and needs good photographs of the animal.