Digital Imagery
and Scanning

ES 555 Small Format
Aerial Photography


Digital imagery is to conventional photographs as digital sound is to conventional audio recordings. In both cases, digital file format offers great advantages for computer handling and processing of the visual or audio data, but with lesser quality compared to traditional analog storage formats. In the case of photographic film, the visual image is captured in minute silver crystals within the photographic emulsion. The ulimate resolution of the image is controlled by the size of these crystals. Recent improvements in digital cameras have achieved the spatial resolution equivalent to standard 35-mm film. New, high-resolution digital cameras equal 70-mm film and larger formats. Thus, digital photography has replaced film for most users.

There is a great need for digital imagery for modern information technology. This need is driving rapid developments in digital cameras as well as scanners for converting conventional film photographs into digital format. Your instructor began using digital cameras in 2001 for small-format aerial photography. In 2004, he discontinued film cameras, and he has employed digital cameras exclusively since 2005. The following table presents some advantages and disadvantages of the methods as utilized for small-format aerial photography.

Conventional film photography vs. digital imagery
for small-format aerial photography
Method Advantages Disadvantages
Conventional
Photography
  • Convenient & inexpensive
  • UV, visible, & near-IR spectrum
  • Superb resolution
  • Easy to archive & sort
  • Dependable under field conditions
  • Delay for developing film
  • Heat sensitive (color)
  • Non-digital imagery
  • Deterioration in storage
    • Digital
    Imagery
  • Convenient & inexpensive
  • Excellent resolution
  • Visible spectrum
  • Images ready "instantly"
  • Easy digital image enhancement
  • Near-IR possible, expensive
  • Changing archive formats


    • Digital image scale

    Scale is a fundamental property of routine aerial photographs, and is especially important for vertical airphotos used for photogrammetric purposes. Interpretability of aerial photographs is often determined by photo scale. However, digital images do not have scale in the conventional sense. A digital image can be viewed and printed in many different scales. Thus, scale becomes a property of the device used to display or print the image file and is not an intrinsic property of the image file itself. In the case of digital imagery, ground sample distance (GSD) is more appropriate as a measure for image resolution. However, GSDs can be quite different for collection, display and products of imagery from the same source. To help understand this somewhat confusing situation, we will examine different types of GSD (based on Comer et al. 1998).

    Collection GSD

    In conventional aerial photography, scale of the photo is equivalent to camera scale, defined as follows--see photogrammetry lecture.

    camera scale = (lens focal length) ÷ (height above ground)

    Consider, for example, a vertical kite aerial photograph taken with a 35 mm lens at a height of 100 m above the ground. The scale would be 0.035 ÷ 100 = 0.00035 (or approximately 1:2860). Note: for this and subsequent calculations, all values are converted into meters.

    Now consider a digital camera with a charged-couple device (CCD). Collection GSD is related to the size of each pixel element within the CCD array as follows.

    collection GSD = (pixel element size) x (height above ground) ÷ (focal length)

    Continuing with the previous example, given a pixel element size of 0.009 mm, then collection GSD would be 0.000009 x 100 ÷ 0.035 = 0.026 meter (about 2½ cm or 1 inch). The digital image "raw" scale is a ratio of CCD pixel element size to collection GSD, in this case 0.000009 to 0.026, or about 1:2888. At this stage, the digital image scale closely approximates the camera scale.

    Display and print scale

    Digital images rarely, if ever, are displayed at the original camera scale, however, which would be much too small for normal visual examination. Most usually, digital images are displayed on a computer monitor, in which the dot pitch controls the image size and resolution, assuming one image pixel is displayed for each monitor dot. In this case, the display scale is a ratio of the collection GSD to the monitor dot pitch.

    display scale = (monitor dot pitch) ÷ (collection GSD)

    Consider a typical monitor with a dot pitch of 0.26 mm and the previous example of collection GSD of 0.026 meter. The display scale would be 0.00026 ÷ 0.026 = 0.01 (or 1:100). A similar calculation can be done for printed digital images. The nominal pixel size for a printer with 300 dpi (dots per inch) is about 0.085 mm. In this case, printer scale would be 0.000085 ÷ 0.026 = 0.00327 (or about 1:300). These examples demonstrate that display and printer scales are usually many times larger than is original digital image scale, because the display pixels are many times larger than the CCD pixel elements.

    The larger scales employed for display and printing of digital images do not imply more information or higher interpretability, however, compared to the "raw" image data. A pixel value is only a pixel value, regardless of the scale at which the pixel is displayed. For visual identification of distinct objects, generally a group of 4 to 9 pixels is necessary (Comer et al. 1998)--see image interpretation.

    Scanning analog photographs

    Digital cameras are rapidly replacing conventional film cameras for all types of aerial photography, as noted above. However, many older film images exist, for which digital versions would be useful. So digital and film will exist side by side for applications in aerial photography for some time to come. In order to exploit the capabilities for computer enhancement and analysis of digital images, it is necessary to convert film (or paper print) photographs into digital format. Two kinds of scanners are in widespread use today for conversion of photographs into digital image files.

    Reference


    Return to SFAP schedule.
    ES 555 © J.S. Aber (2008).