CCD Read Noise Summary
Intro l
Measure l
Interpret FFTs l
Noise FFTs l
Your Camera l
Create FFTs l
Summary
Summary
Achieving high dynamic range, low noise images from a cooled CCD camera requires an understanding of how CCDs work and the different sources of noise that can reduce the quality of your images. Those noise sources include:
| | • |
Dark Current |
| | • |
Pixel Non-Uniformity |
| | • |
Shot Noise |
| | • |
On-Chip CCD Read Noise |
| | • |
Off-Chip CCD Camera Noise |
| | • |
Electronic Interference |
Some of these sources of noise, such as dark current and pixel non-uniformity, can be reduced with standard image calibration techniques using dark frames, flat fields and bias frames. Noise sources that display a Gaussian distribution, such as shot noise and on-chip CCD read noise can be smoothed out by average or median combining multiple images. Other sources of noise, such as periodic noise introduced by the camera’s electronics are more difficult to deal with because these types of noise build up just like the signal from the stars you’re trying to image.
We presented techniques to identify and measure the read noise in CCD cameras:
| | • |
Examine bias frame histograms |
| | • |
Isolate camera read noise by creating “Read Noise Frames” from multiple bias frames |
| | • |
Use Fast Fourier Transforms (FFT) to identify non-random, periodic noise |
| | • |
Visually interpret Read Noise Frame FFTs |
Lastly we provided instructions demonstrating how to collect the images necessary to isolate the read noise in any CCD camera and then produce FFTs of the read noise frames to identify any periodic noise in those images.
| Digital image processing techniques can often produce excellent results even with source images that have a reduced dynamic range and substantial amounts of read noise, but whether you’re producing detailed photometric data for a university research project or just want to produce great pictures of your favorite deep sky targets, producing the best images always begins with the best data. |
Downloads
You can download the original, QSI bias frames used for these examples by clicking the link below. The zip file contains the following:
3 QSI 504 bias frames (768x512, FITS format)
1 QSI 504 read noise frame (512x512 FITS format) [Bias1 – average of 3 bias frames]
1 FFT of QSI 504 read noise frame (512x512 TIFF format):
Download QSI 504 read noise examples (.ZIP 1.05mb)
References
The easy-to-understand visual FFT examples were inspired by, “An Intuitive Explanation of Fourier Theory” by Dr. Steven Lehar. If you’re interested in gaining a deeper understanding of how Fourier theory can be used for image processing, Dr. Lehar’s essay is an excellent place to start.
If you have questions about CCD camera read noise or Fourier transforms as they relate to scientific cameras and image processing, please write to us using our contact web form. We generally respond to email questions within 24 hours.
<< Return to CCD Read Noise Introduction
|
