A Treasure Trove of Nature: The advances and challenges of digitising natural history specimens

Presentation given by Steen Dupont and Laurence Livermore for the British Computer Society's Advanced Programming Specialist Group, London. 16 May 2019.Abstract. The Natural History Museum holds over 80 million specimens, ranging from tiny prehistoric shark teeth, massive meteorites and some of the most comprehensive collections of contemporary animals and plants. While the museum has an established Digital Collections Programme many challenges still remain in effectively digitising these specimens and extracting, linking and sharing their data. Over the past three years we have developed new semi-automated workflows and experimented with automated image analysis, text mining and machine learning on images of our specimens and related literature. We will discuss some of the physical and digital challenges, our current approaches and plans for the future