In the previous chapter, we outlined how to make support frames—darks, flats, and bias frames—for your images. In this chapter, we examine the nitty-gritty details of image calibration and explain what it is and why it is necessary.
Images straight from the CCD carry a number of unwanted "signals." The goal of calibration is to correct the raw image so that the calibrated (but otherwise unprocessed) image contains pixel values that accurately portray the intensity of light that fell on the CCD during exposure.
The unwanted signals in a raw CCD image include two additive components and one multiplicative component. The additive components are a voltage offset, or bias, from zero volts; and a signal generated by thermal emission of electrons that grows linearly with exposure time. The multiplicative error arises because photosites have differing sensitivities to light. Calibration involves removing the bias, subtracting the dark current, and dividing the image by a map of photosite sensitivity.
Performing these corrections exacts a toll on the CCD user: the bias, dark current, and sensitivity map of the CCD must be determined. Ideally, these should be determined one or more times during each observing session, but in practice the calibration regimen may be simplified.
This chapter examines the behavior of the unwanted signals found in raw CCD images and suggests three observational protocols—basic, standard, and advanced—for calibrating CCD images. The word "protocol" means a plan or prescription for accomplishing a task. In this context, calibration protocols are recommendations as to how an observer using an astronomical CCD camera can make and calibrate satisfactory images.
• The basic protocol meets the needs of many observers, removing bias and dark current for a relatively small investment in observing time. During an observing session, from time to time, the observer takes dark frames having the same integration time as the images. To calibrate the latter, the observer combines dark frames to reduce noise and subtracts them from the raw CCD images. Basic calibration is suitable for search and survey images, and often meets the needs of observers who are just getting into CCD observing.
• The standard protocol adds to the basic protocol a correction to compensate for the CCD's photosite-to-photosite sensitivity variation. In addition to taking dark frames, at some point during the night, the observer shoots flat-field frames either from the twilight sky or from a low-intensity illuminated box or panel. Standard calibration produces high-quality monochrome or color images suitable for display or publication, as well as images that can be used for precise astrometry or photometry.
• The advanced protocol is a prescription for producing images of "research" quality, appropriate for use with high-performance scientific-grade CCDs. In addition to dark frames and flat-field frames, the observer obtains bias frames that permit more precise removal of the bias and dark current than does the standard protocol. Advanced calibration methods are also more flexible and forgiving than the basic and standard protocols, and thus often repay the additional time required to obtain the required bias-frame data.
• Tip: AIP4Win supports the basic, standard, and advanced calibration protocols. As an observer, remember that calibration frames must be taken when you make your observations.
From the foregoing, it should be clear that the basic calibration protocol requires the least work. All that's necessary is that during imaging, you shoot an adequate number of dark frames. The biggest constraint is that the integration time you use for your dark frames must be the same as the integration time you are using for your images. Since most observers settle on a convenient integration time, such as 60 seconds, there is no difficulty in making dark frames with the same integration time.
Standard calibration is a step more sophisticated because it corrects for variations in CCD sensitivity and subtracts the dark current. In addition to shooting an adequate number of dark frames, the observer must also shoot a set of flat-field frames and matching flat-field dark frames during the night. Flat-fielding corrects for the dark corners caused by vignetting and for the little dark donuts caused by dust as well as for variations in CCD sensitivity. Most observers will probably find that the standard calibration protocol is entirely adequate.
A step more complex than the standard protocol is the advanced calibration protocol. Within an already crowded shooting schedule, observers must find time to shoot an adequate number of bias frames. However, by doing so, they gain a significant advantage: when a bias frame is subtracted from a dark frame, the resulting image (called a thermal frame) contains only dark current. Because dark current increases linearly with time, a thermal frame can be scaled to match integrations of any length. Thus the observer can shoot a set of dark frames with 300-second integrations. During calibration, a master bias frame is subtracted from the dark frame to produce scalable thermal data. This can be scaled and then subtracted from integrations of 5 seconds, 30 seconds, 150 seconds—any integration shorter than the integration used for the dark frame.
• Tip: For optimum results, decide in advance which calibration protocol you will use; then be sure to obtain the necessary calibration frames during the observing session. Most observers choose either the basic or standard calibration protocol.
Was this article helpful?