Abstract

Bird populations are an important bio-indicator, ; so collecting reliable data is useful for ecologists helping conserve and manage fragile ecosystems. However, existing manual monitoring methods are labour-intensive, time-consuming, and error-prone. The aim of our work is to develop a reliable system, capable of automatically classifying individual bird species in flight from videos. This is challenging, but appropriate for use in the field, since there is often a requirement to identify in flight, rather than when stationary. We present our work in progress which uses combined appearance and motion features to classify and present experimental results across seven species using Normal Bayes classifier with majority voting and achieving a classification rate of 86%.