6.2 Digital imaging

The use of digital cameras in the field is rapidly becoming standard practice on account of the instantly verifiable results, cheap hardware and minimal media costs to store the digital results. The Charge Coupled Device (CCD) employed within digital cameras that records the photograph is a great deal more sensitive (about ten times) than conventional film. This makes these devices ideal for use in archaeological fieldwork, where we are very often concerned with documenting very slight variations in soil texture and colour. The ease with which digital images can be enhanced using Adobe Photoshop or similar software has made many of our darkrooms, with their wet-processing areas and complex enlargers, obsolete. With semi-professional digital cameras now selling for less than £1000, with all the flexibility of traditional Single Lens Reflex (SLR) cameras with interchangeable lenses and a multitude of different operating modes, the digital camera has come of age.

Two factors have potentially posed problems for applications related to excavation; the lack of very wide-angle lenses suitable for use with digital cameras, and the number of pixels or resolution recorded by the CCD. Since the CCD captures light in RGB colours (Red, Green and Blue) it must be remembered that the megapixel values are multiplied by 3 to calculate the file size, thus a 5 megapixel camera will generate 15 megabyte files and as resolution increases the increasing file size has an impact on the time taken to save images to the storage media in the camera. This has a bearing, for instance, in air photography where one wishes to photograph multiple frames at a high frequency.

In most cases the CCD is about half the size of a 35mm negative. Full frame cameras are now being released but at present remain expensive; this reduction in the picture area has meant that the effective focal length of the lens is almost doubled when compared to a conventional 35mm camera. The resolution (expressed in megapixels) is based on the number of elements in the CCD for recording the image. This is now less of an issue, with domestic 5 megapixel cameras now available for less than £200. If we were to take a standard 5 megapixel image with a resolution of 2560 x 1920 pixels and print it out at full resolution on a full colour printing press the resultant image would be 21.6 x 16.3cm (8.5 x 6.4 inches); almost all printing presses operating at 300 dots per inch. Regardless of the resolution provided to a printer, the print resolution will not change and a preferred image resolution of 304dpi will give perfect results. Higher resolutions offer the potential to produce even larger printouts or the ability to produce high-resolution images of portions of the frame. They may also produce better results using ink-jet printers, where very high resolutions are required to ensure good colour or tonal range. Perhaps more critical than simple pixel resolution is the format of the saved images. Most cameras now generate RAW, TIF and JPEG images. RAW format saves the actual data recorded by the CCD and offers the maximum post-processing potential. TIF files offer the second-best result; as they are pre-processed within the camera, the TIF files produce images at full resolution but with less information than in the RAW files. JPEG files maximise the use of storage space but employ 'lossy' image compression techniques to reduce the file size. JPEG format files should be avoided if possible as they compromise the actual image data.

In the case of the DigIT project we used two main digital cameras, a Sony DS F505v (3.6 megapixel) and a Fuji S1 PRO, SLR (3.5 megapixel CCD interpolated using sophisticated techniques on the CCD to 6 megapixel). The results from these two digital cameras were almost indistinguishable at normal print sizes, although the Fuji S1, by virtue of its through-the-lens viewfinder, exchangeable lenses and more sophisticated imaging options, was by far the more flexible. While neither of the digital cameras matched the 6 x 6cm Hasselblad, they were entirely fit for purpose and the results can be seen throughout this article.

With camera resolution and storage media capacities rapidly increasing, the ease with which gigabytes of digital images can be gathered is worrying. Although the database requirements incorporating the image-associated metadata per photographic frame are no different from that required to document images recorded on film, the sheer ease with which digital photographs can be accumulated greatly increases the necessary record keeping. The simple process of renaming 1000 image files to something more useful than a number takes time and requires the use of a well-structured directory tree if chaos is not to be the end result. Although it is possible to incorporate imagery within conventional databases such as Access, this is not good practice. A far more useful approach is to use a directory tree to manage the primary image files and use a URL within the database record, documenting each image to point to the image.

The singular problem arising from the use of digital photography is the long-term survival of the results. Conventional film, if properly processed and stored, should last well over 100 years. In contrast, the long-term storage of digital images will rely on the storage media, its storage, and the ability of future software to interpret the results. Writeable CDs and DVDs are ideal for short-term storage, although they are subject to permanent damage and data loss if left in the sun (even in boxes where only the edge receives sunlight). Data storage costs reduce at a rate that closely matches our ability to fill even more space, and it is simply not possible to guarantee long-term survival of these data. Mega-data stores, such as that provided by the Archaeology Data Service, dedicated towards securing long-term archiving of digital archaeological data, will not be able to provide permanent storage for the terabytes of digital photographs that British archaeology could generate in a year. We may have to adopt a pragmatic solution and preserve important images as archival prints, requiring the use of special inks and paper which have a specified survival time in excess of 100+ years, if we wish to guarantee the survival of our digital image archive. The need for archive-quality prints adds to the excavation costs, a factor which should be considered by anyone contemplating the use of digital photography as the sole photographic record from an excavation.

The opportunities that digital photographic recording offers are as yet untapped and will change as the technology continues to evolve. For example, time-lapse photography of the excavation of complex deposits such as furnished graves can be tremendously helpful at the post-excavation stage but in the past was very difficult to implement. This is now a supported function on many digital cameras.

In addition to documenting the excavation through conventional and digital still photography, digital video was taken to record the excavation process and for the preparation of materials for distribution on the World Wide Web. Low-resolution MPEG files were created using the Sony DSC-F505V digital camera; better quality digital video was captured using a Sony DCR-PC100E Digital Video camera and archived on analogue tape. The latter proved difficult to convert into suitable formats for use on the Web and required very large amounts of hard-disk space; thus it was simply documented and archived for future reference.


© Internet Archaeology/Author(s) URL:
Last updated: Wed Nov 11 2009