Report on trial of SatScan tray scanner system by SmartDrive Ltd.

Wed, 2012-06-13 09:45 -- kevinlove
TitleReport on trial of SatScan tray scanner system by SmartDrive Ltd.
Publication TypeReport
Year of Publication2010
AuthorsBlagoderov, V., Kitching I., Simonsen T., and Smith V.
InstitutionNatural History Museum, London
KeywordsImaging, Technology
URLhttp://vsmith.info/files/npre20104486-1.pdf
Description: 
Smartdrive Ltd. has developed a prototype imaging system, SatScan, that captures digitised images of large areas while keeping smaller objects in focus at very high resolution. The system was set up in the Sackler Biodiversity Imaging laboratory of Natural History Museum on March 8, 2010 for a one-month trial. A series of projects imaging parts of the entomological, botanical, and palaeoentomological collections were conducted to assess the system's utility for museum collection management and biodiversity research. The technical and practical limitations of the system were investigated as part of this process.The SatScan system facilitates the capturing of a very large number of good quality images in a very short time. Large parts of the NHM collection could be digitised in dorsal view extremely quickly. These images have a wide range of uses across research, collection management, and public engagement. Scalability of the system is limited by our desire to assign unique identifiers (a number and/or a barcode) to specimens, and the cropping of these images. Without these identifiers digitised images will have limited long term value. The assignment of specimen level identifiers is potentially labour intensive. Options for assigning identifiers were not investigated as part of this trail but include the use of physical labels on each specimen (with significant resource implications and a significant volunteer workforce) and the use of virtual identifiers (as a virtual layer over the image, perhaps automatically assigned, and possible coupled with physical labels attached to specimens as dictated by recuration activities). Intuitive software (with a web interface) needs to be developed to facilitate this process, including support for cropping of an image and the automatic assignment and printing of an identifier label. On-demand assignment of identifiers would allow us to prioritize the digitisation but it will represent a significant change to the way we curate our collections and would require sustained and ongoing support from Collection Management. Additional concerns relate to the amount of storage space required to manage images, connection with existing digital systems and the utility of dorsal images for certain parts of the collection. These problems need to be addressed as part of a larger scale study.