Ornithology meets the IoT: Automatic Bird Identification, Census, and Localization

Stopping the unprecedented decline of biodiversity is one of the biggest environmental challenges in the 21
^st
century. Regional monitoring of selected bird species is one key to analyze the causality and detect changes because their presence is a good indicator of ecosystem health and integrity. To enable cost-efficient, long-term monitoring, this paper presents a Wireless Acoustic Sensor System (WASS) for automated remote bird identification, census, and localization. The proposed system is based on a Wireless Acoustic Sensor Network (WASN) to record and transmit the audio samples combined with a classification framework for automated evaluation. This paper does not only discuss the technical requirements and introduces a user-friendly system architecture but also presents a first prototype. Its functionality is assessed in a short outdoor deployment. The measurements prove that the proposed system is capable of monitoring and classifying the present species automatically. Still, there are some limitations, and future refinements are pointed out.