An analysis of CCD camera noise and its effect on pressure sensitive paint instrumentation system signal-to-noise ratio

Abstract

Quantitative pressure measurements can be acquired by pressure sensitive paint (PSP) instrumentation systems incorporating charge-coupled devices (CCD) for PSP photoluminescence image detection. However, intrinsic CCD noise corrupts the PSP image, manifesting in erroneous measurements of pressure and the corresponding coefficient of pressure (C/sub p/). This manifestation is analyzed and quantified in terms of PSP image signal-to-noise ratio (SNR). The image acquisition and processing requirements maximizing a PSP image's SNR are determined. Specific attention is given to the PSP image calibration or flat fielding for CCD scene noise. This algebraic calibration is standard practice in quantitative CCD imaging. However, when applied to PSP images, subtleties arise due to the relative motion between wind-on and wind-off images used in the PSP ratio method. By averaging images and calibrating them for scene noise before ratioing, these subtleties are addressed and the following are true. The PSP C/sub p/ uncertainty will be determined by the image least averaged. Therefore, the mean flat field image must be based on 5 and 50 times more samples than the mean raw PSP image to be within 10% and 1% of the theoretically minimum uncertainty, respectively.