Capture Control Of Digital Cameras

H. G. Dietz
http://aggregate.org/hankd/

Department of Electrical and Computer Engineering
Center for Visualization & Virtual Environments
University of Kentucky, Lexington, KY 40506-0046

Initial posting: February 3, 2007

Latest update: June 3, 2007

This document (which currently spans four URLs) should be cited using something like the bibtex entry:

@techreport{capture20070602,
author={Henry Gordon Dietz},
title={Capture Control Of Digital Cameras},
year={2007},
month={June},
institution={University of Kentucky},
howpublished={Aggregate.Org online technical report},
URL={http://aggregate.org/DIT/CAPTURE/}
}

Before there was digital still photography, "remote control" of a camera had a rather unambiguous meaning: it meant that there was a cable release or an electronic control (wired or wireless) that allowed the user to take a picture without physically touching the camera. As digital cameras developed, two entirely separate types of "remote control" of cameras became possible:

1. Downloading or uploading of pictures with a host computer
2. Use of a host computer to tell the camera to take a picture

The ability to download pictures from a camera to a host computer was immediately critical because there wasn't much else one could do with a digital image. Thus, even early digital cameras like the Casio QV10 had such support.

Tethered control of image capture generally has been treated as a specialized feature used primarily for scientific (astro and micro photography) and certain types of studio photography. However, the technique has become very popular for kite aerial photography (often called KAP -- see KAPer e-magazine), and really has very wide applicability. For example, despite what one sees in TV shows and movies, surveilance camera video is generally of very poor quality. Using a tethered compact digital still camera in a security application not only allows capture of low-quality video using image preview support, but also capture of one or more high-resolution (perhaps even using flash) images when the video stream shows something potentially of interest. We use tethered still cameras in applications ranging from autonomous vehicles to digitizing 3D scenes.

For example, the following slow-update live image is from a tethered Nikon 950 digital still camera with a 185 degree fisheye lens:

At this writing, that one camera has captured over 2 million images under computer control... and it is still capturing nice images at its full native 2MP resolution. We have Olympus D340R cameras that have taken even more tethered images than that Nikon, although the infinity focus stop slowly shifted over time requiring macro focus mode to be set in order to focus at normal distances. There are lots of other compact digital cameras that we've used less extensively but with even better results. Cheap DSLRs (digital single-lens reflex cameras) use larger sensors that can provide even better image quality, although the relatively high mechanical complexity of DSLRs will probably give them shorter lifespans. In any case, the quality and flexibility of commodity digital cameras under computer control is really shockingly good at a very low cost.

We have decomposed the tethered control problem into three separate aspects:

• Hardware Modifications For Triggering Capture
  For cameras that do not directly support computer control, here are some ways to at least trigger the shutter.
• Software For Tethered Capture Control
  Given some type of hardware support, this is a summary of the software available to implement the computer control of capture and perhaps various camera parameters.
• Camera Pan/Tilt (Motion) Control
  Most cameras do not directly have pointing control, but here is a brief discussion of how it can be implemented.

Note that, before you begin to implement computer-controlled capture mechanisms, you might need to resolve potentially serious legality issues involving the need for consent. Audio recording is typically the primary legal concern, but some states have laws that also limit recording of images. A good overview of the issues is given by the The Reporters Committee for Freedom of the Press. Note that what you intend to use the images for can make a difference in legality. Beyond that, there are strange issues such as, in the case of a "conversation" between a person and a computer, can the computer be the party to the conversation that is aware the conversation is being recorded? For that matter, does computer processing of such data imply that a recording was made even if the raw data is not archived?

Our solution has been to clearly mark each area that might be subject to computer processing of captured images, audio, or any other sensor data with a notice to that effect. Typically, we post a notice sticker in an obvious place at each entry to each such area, and we're fairly comfortable that this meets Kentucky's requirements given our research purpose. Don't blindly do what we did -- check your local laws.

The only thing set in stone is our name.