Reevaluating digital vs. conventional photographic cameras for research


Reevaluating Digital vs. Conventional Photographic Cameras for Research

Gilbert and Richards (2000) have performed a service by pointing out that visible light photography, especially when enhanced by digital manipulation, offers new capabilities for data analysis. Such widely used programs as Adobe Photoshop™ make it practical to enhance data that are present but perhaps not terribly obvious in photographs. I have used some of these “tricks” myself to make the patterns of muscle fibers more apparent, and to analyze spatial relationships among them (Ledoux et al., 2001).

I disagree, however, with their statements that digital cameras are the way to go. They are not necessary, and it can be argued that they are not an appropriate choice when the best quality images are needed. Film cameras have the advantages over digital cameras of higher resolution, greater color depth, and a greater tonal range. This will not always be true but it is now, especially with low-resolution digital cameras. One rule of thumb I have seen is that a 6-megapixel camera is needed to produce an 8“ × 10” print with detail comparable to a film image, and that ignores questions of color and shading.

When great detail and subtlety are not called for, digital cameras do have the advantage of allowing results to be obtained quickly, and, in fact, the quality may be more than adequate for the sort of work Gilbert and Richards are doing.

Another factor is cost. A high-end digital camera is expensive. The Nikon™ D1 Gilbert and Richards use produces images of about 2.7 megapixels, and sells for over $4,000 without a lens. Extremely high-quality film cameras are available for a small fraction of that price. In addition, 35-mm film can be converted to digital images very easily and inexpensively. For example, the Kodak PhotoCD™ is a system that converts slide or negative images into digital form at resolutions up to about 6 megapixels (or about 25 megapixels if needed, at higher cost). The cost per slide at a professional photo lab is about $2.00. One PhotoCD holds about 100 images and the images can be imported into such programs as Photoshop for manipulation. The lab I use has a standard two-day turnaround, and offers 24-hour service as an option.

There is one more factor that might be important in some cases: permanence. Digital formats and media type seem stable over the short range, but may not be over longer periods. A case in point: a few years ago, Syquest™ produced removable disks that were an industry standard; today they are essentially out of business. There is talk of CDs being replaced by higher capacity media similar to DVDs. Five years or so ago, 10“ tape reels were common, but today I wouldn't know where to find a tape reader for them. Also, there is limited information on the long-term stability of the media themselves, while film can be stored for decades with little or no degradation.

The authors claim that it is possible to use a smaller f-stop with digital cameras than with film cameras, thereby increasing the depth of field. I think that in most laboratory situations, except perhaps when conditions limit the amount of light that can be used, it is rather simple to adjust the lighting to allow a smaller f-stop. Some systems allow flash photography with through-the-lens control of the flash, which has the additional advantage of eliminating motion artifacts when camera and subject cannot be rigidly fixed. It is also worth noting that at high magnifications, using a diaphragm to increase depth of field degrades image resolution because of diffraction effects.

Finally, I want to correct one small error. C-mount lenses are not regular 35-mm camera lenses. They are most commonly used in 16-mm cine and video cameras, and cover a film or CCD area of 0.5–0.67 inches. This is adequate for the CCDs used in most digital cameras, because they are usually much smaller than a 35-mm film frame, and adapters for the lenses are available.

I am glad to see that The New Anatomist published an article on the use of digital photography in expanding the scientific value of images, but I also think that it is important to note that there are different ways of doing so.