View Item 
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        JavaScript is disabled for your browser. Some features of this site may not work without it.

        Browse

        All of UU Student Theses RepositoryBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

        Preserving biodiversity by identifying individuals within species with the help of artificial intelligence

        Thumbnail
        View/Open
        Bachelor_Thesis_Artificial_Intelligence_Sander_Kaspers_5744512.pdf (8.719Mb)
        Publication date
        2021
        Author
        Kaspers, S.K.
        Metadata
        Show full item record
        Summary
        As the number of endangered species in the world increases due to human activity, biodiversity is decreasing rapidly. Obtaining knowledge about endangered species is the key to stopping this decline. Artificial intelligence is used to speed up the process of analyzing camera trap data. Identifying individuals of endangered species can offer insights in population numbers and movement patterns. Automatic detection and classification of species is necessary to obtain a dataset that contains a certain endangered species. Such a dataset provides the opportunity to identify individuals within species. This study analyzes the artificial intelligence techniques that can be used for both of these tasks. At this moment, HotSpotter and StripeSpotter are two prominent types of software that can be used for identifying individuals within species. To test which of the two performs best, I applied both types of software to a self-made dataset of snow leopard individuals. Regarding biodiversity, it is important to obtain knowledge about snow leopard populations. Because the snow leopard is an endangered species that as a rare predator is of great importance for preserving biodiversity, I chose to test both types of software on snow leopard individuals. I tested both types of software also on a dataset of zebra individuals to verify my hypothesis. The results show that HotSpotter outperforms StripeSpotter for both datasets. For this reason, my conclusion is that HotSpotter can be used best for identifying individuals within species.
        URI
        https://studenttheses.uu.nl/handle/20.500.12932/40648
        Collections
        • Theses
        Utrecht university logo