Anzeige
Erschienen in:
2018 | OriginalPaper | Buchkapitel
Automated Animal Recognition Platform
verfasst von : Bastian Hermann, Jürgen Sieck
Erschienen in: Digitisation of Culture: Namibian and International Perspectives
Verlag: Springer Singapore
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
This paper describes the conception and implementation of a web-based platform which is able to recognise wildlife animal species based on user-provided images. The service is designed for tourists and non-professional biologists who like to observe animals in their natural habitat. This paper focuses on a new approach to determine animal species based on pattern and colour features. As a consequence of the specific user context, the service makes use of a combination of both recognition models implemented as a cross-validation. We showed that the pattern recognition based on local binary pattern histograms achieves good results on a database with images of different animals. On the other hand, this paper examined different colour models for the colour recognition model for this specific context and setup, in which the HSV colour model achieved overall the best results.