Diatom Identification: A Double Challenge Called ADIAC

This paper introduces the project ADIAC (Automatic Diatom Identiication and Classiication), which started in May 1998 and which is nanced by the European MAST (Marine Science and Technology) programme. The main goal is to develop algorithms for an automatic identiication of diatoms using image information: both valve shape (contour) and ornamentation. The paper presents the goals of the project as well as rst results on shape modeling and contour extraction. Public data are available in order to create student projects beyond the ADIAC partnership. For further information see 1 Introduction ADIAC is the acronym of the project Automatic Di-atom Identiication and Classiication. In phycological research the word identiication refers to what in pattern recognition is meant by classiication, whereas the phycological meaning of classiication is the establishment of the class-forming rules. In order to avoid a confusion, we will apply the phycological interpretation. Since this is the rst scientiic publication by most of the project's partners, and because our goal is to promote diatom identiication as a new and challenging area in pattern recognition, we start with explaining the history of the project, diatom research and the goals. Sections 2 to 4 present rst results on image databases, shape modeling and contour extraction.