Jump to content Jump to navigation

Media library

Naturalis is busy digitizing its entire collection of zoological and botanical specimens.

All of the specimens are being registered in a central collection registration system, and from each specimen one or more high-resolution photos are being made.

The aim of this project is to make the Naturalis collection digitally available to scientists, amateurs and other interested persons throughout the world. The project covers a total of 7 million collection objects and 4 million of these have already been processed.

Photos are made in so-called digistreets. For example there are digistreets for insects, vertebrates, fossils, glass slides, etc. All of the digistreets together produce up to 4 terabytes (4000 gigabytes) in photos per day. All of these photos are registered in the media database and a backup is made each of them, which ultimately ends up at the Nederlands Instituut voor Beeld en Geluid (Netherlands Institute for Sound and Vision). Furthermore, several smaller versions of each photo must be made.

To ensure everything proceeds as planned, the department of software development at Naturalis has developed special software that is internally known as the Media Library. The Media Library is capable of processing enormous quantities of photos in parallel. Not only are the daily productions of the different digistreets processed in parallel, but the individual tasks within the processing process (indexing, making a backup, reduction) take place in parallel as well. This enables Naturalis to process several terabytes per day and to publish these on the web with the use of relatively few servers.

The following step for Naturalis is the development of a search application for the online photo material. Although the Media Library can be used to request photos of a collection object online, this only works if you know the barcode that has been allocated to the object. With the development of the Media Library Search Facility, users will be able to search for images based on characteristics such as the species name of the specimen.