Flat Field Correction

http://www.roper.co.jp/Html/roper/tech_note/html/tflatfield.htm

A CCD imager is composed of a two dimensional array of light sensitive detectors or pixels. The CCD array is mechanically quite stable with the pixels retaining a rigidly fixed geometric relationship. Each pixel within the array, however, has its own unique light sensitivity characteristics. As these characteristics affect camera performance, they must be removed through calibration. The process by which a CCD camera is calibrated is known as "Flat Fielding" or "Shading Correction".

Flat fielding can be illustrated in the following equation:

IC = [(IR - IB) * M] / (IF - IB)

Where IC is the calibrated image; IR is the non-calibrated object exposure; IB is the bias or dark frame; M is the average pixel value of the corrected flat field frame; and IF is the flat field frame.

IB Flat fielding requires the acquisition of two calibration frames. First, a bias frame or a dark frame should be taken. Bias clears the camera of any accumulated charge and reads out the cleared CCD. The resulting image is a low signal value image. In this image, all of the pixels have approximately the same value, which consists of the electronic offset of the system of the inherent structure of the CCD. Dark clears the CCD of charge, allows charge to accumulate for a specified amount of time with the shutter closed and then reads out that charge (dark current). A dark frame contains the standard bias component as well as the dark signal. The dark command is most useful when taking long exposures with low light levels.

IF The second calibation image, the flat field frame, measures the response of each pixel in the CCD array to illumination and is used to correct for any variation in illumination over the field of the array. The optical system most likely introduces some variation in the illumination pattern over the field of the array. The flat fielding process corrects for uneven illumination, if that illumination is a stable characteristic of each object exposure. Thus, it is necessary to illuminate the CCD with a light pattern that is as representative of the background illumination as possible. This illumination should be bright enough, or the exposure made long enough, so that the CCD pixels signals are at least 25 percent full scale or preferably higher. For a Photometrics camera equipped with a 12 bit analog processing card, the level should be at least 1000 ADUs.

IR An exposure of the object of interest is acquired.

(IR - IB) The object frame must be corrected for electronic offset by subtraction of the bias/dark frame from it.

(IF - IB) The flat field frame must also be corrected for electronic offset by subtraction of the bias/dark frame from it. The average pixel value of the bias/dark corrected flat field frame must then be ascertained (M).