Signal-to-noise ratio (imaging)

The signal-to-noise ratio (SNR) is used in imaging as a physical measure of the sensitivity of a (digital or film) imaging system. Industry standards measure SNR in decibels (dB) of power and therefore apply the 10 log rule to the "pure" SNR ratio (a ratio of 1:1 yields 0 decibels, for instance). In turn, yielding the "sensitivity." Industry standards measure and define sensitivity in terms of the ISO film speed equivalent; SNR:32.04 dB = excellent image quality and SNR:20 dB = acceptable image quality.^[1]

Definition of SNR

An operator arbitrarily defines a box area in the signal and background regions of a back-illuminated half moon or knife-edge test target. The data, (such as pixel intensity), is used to determine the average signal and background values.

Traditionally, SNR has been defined as the ratio of the average signal value $\mu _{{\mathrm {sig}}}$ to the standard deviation $\sigma _{{\mathrm {bg}}}$ of the background:

{\mathrm {SNR}}={\frac {\mu _{{\mathrm {sig}}}}{\sigma _{{\mathrm {bg}}}}}

However, when presented with a high-contrast scene, many imaging systems clamp the background to uniform black, forcing $\sigma _{{\mathrm {bg}}}$ to zero, artificially making the SNR infinite.^[2] In this case a better definition of SNR is the ratio of the average signal value $\mu _{{\mathrm {sig}}}$ to the standard deviation of the signal $\sigma _{{\mathrm {sig}}}$ :

{\mathrm {SNR}}={\frac {\mu _{{\mathrm {sig}}}}{\sigma _{{\mathrm {sig}}}}}

which gives a meaningful result in the presence of clamping.

Calculations

Explanation

The line data is gathered from the arbitrarily defined signal and background regions and input into an array (refer to image to the right). To calculate the average signal and background values, a second order polynomial is fitted to the array of line data and subtracted from the original array line data. This is done to remove any trends. Finding the mean of this data yields the average signal and background values. The net signal is calculated from the difference of the average signal and background values. The RMS or root mean square noise is defined from the signal region. Finally, SNR is determined as the ratio of the net signal to the RMS noise.

Polynomial and coefficients

The second order polynomial is calculated by the following double summation.

$f_{i}=\sum _{{j=0}}^{m}\sum _{{i=1}}^{n}a_{j}x_{i}^{j}$

- $f\,$ = output sequence
- $m\,$ = the polynomial order
- $x\,$ = the input sequence (array/line values) from the signal region or background region, respectively.
- $n\,$ = the number of lines
- $a_{j}\,$ = the polynomial fit coefficients
The polynomial fit coefficients can thus be calculated by a system of equations.^[3]

${\begin{bmatrix}1&x_{1}&x_{1}^{2}\\1&x_{2}&x_{2}^{2}\\\vdots &\vdots &\vdots \\1&x_{n}&x_{n}^{2}\end{bmatrix}}{\begin{bmatrix}a_{2}\\a_{1}\\a_{0}\\\end{bmatrix}}={\begin{bmatrix}f_{1}\\f_{2}\\\vdots \\f_{n}\end{bmatrix}}$

Which can be written...

${\begin{bmatrix}n&\sum x_{i}&\sum x_{i}^{2}\\\sum x_{i}&\sum x_{i}^{2}&\sum x_{i}^{3}\\\sum x_{i}^{2}&\sum x_{i}^{3}&\sum x_{i}^{4}\end{bmatrix}}{\begin{bmatrix}a_{2}\\a_{1}\\a_{0}\end{bmatrix}}={\begin{bmatrix}\sum f_{i}\\\sum f_{i}x_{i}\\\sum f_{i}x_{i}^{2}\end{bmatrix}}$

Computer software or rigorous row operations will solve for the coefficients.

Net signal, signal, and background

The second-order polynomial is subtracted from the original data to remove any trends and then averaged. This yields the signal and background values:

\mu _{{\text{sig}}}={\frac {\sum _{{i=1}}^{n}(X_{i}-f_{i})}{n}}\qquad \qquad \mu _{{\text{bkg}}}={\frac {\sum _{{i=1}}^{n}(X_{i}-f_{i})}{n}}

where

$\mu _{{\text{sig}}}$ = average signal value
$\mu _{{\text{bkg}}}$ = average background value
$n\,$ = number of lines in background or signal region
$X_{i}\,$ = value of the i^th line in the signal region or background region, respectively.
$f_{i}\,$ = value of the i^th output of the second order polynomial.

Hence, the net signal value is determined by :

{\text{signal}}=\mu _{{\text{sig}}}-\mu _{{\text{bkg}}}

RMS noise and SNR

The RMS Noise is defined as the square root of the mean of variances from the background region.^[2]

{\text{RMS noise}}={\sqrt {{\frac {\sum _{{i=1}}^{n}(X_{i}-{\frac {\sum _{{i=1}}^{n}X_{i}}{n}})^{2}}{n}}}}

The SNR is thus given by

{\text{SNR}}={\frac {{\text{signal}}}{{\text{RMS noise}}}}

Using the industry standard 20 log rule^[4]...

{\text{SNR}}=20\log _{{10}}{\frac {{\text{signal}}}{{\text{RMS noise}}}}\,{\mbox{dB}}

References

↑ ISO 12232: 1997 Photography – Electronic Still Picture Cameras – Determining ISO Speed here
1 2 Mazzetta, J.; Caudle, Dennis; Wageneck, Bob (2005). "Digital Camera Imaging Evaluation" (PDF). Electro Optical Industries. p. 8. Retrieved 28 March 2010.
↑ Aboufadel, E.F., Goldberg, J.L., Potter, M.C. (2005).Advanced Engineering Mathematics (3rd ed.).New York, New York: Oxford University Press
↑ Test and Measurement World (2008). SNR. In Glossary and Abbreviations.http://www.tmworld.com/info/CA6436814.html?q=SNR

Noise (physics and telecommunications)

General	Acoustic quieting Distortion Noise cancellation Noise control Noise measurement Noise power Noise reduction Noise temperature Phase distortion

Noise in...	Audio Buildings Electronics Environment Government regulation Human health Images Radio Rooms Ships Sound masking Transportation Video

Class of noise	Additive white Gaussian noise (AWGN) Atmospheric noise Background noise Brownian noise Burst noise Cosmic noise Flicker noise Gaussian noise Grey noise Jitter Johnson–Nyquist noise (thermal noise) Pink noise Quantization error (or q. noise) Shot noise White noise Coherent noise Value noise Gradient noise Worley noise

Engineering terms	Channel noise level Circuit noise level Effective input noise temperature Equivalent noise resistance Equivalent pulse code modulation noise Impulse noise (audio) Noise figure Noise floor Noise shaping Noise spectral density Noise, vibration, and harshness (NVH) Phase noise Pseudorandom noise Statistical noise

Ratios	Carrier-to-noise ratio (C/N) Carrier-to-receiver noise density (C/kT) dBrnC E_b/N₀ (energy per bit to noise density) E_s/N₀ (energy per symbol to noise density) Modulation error ratio (MER) Signal, noise and distortion (SINAD) Signal-to-interference ratio (S/I) Signal-to-noise ratio (S/N, SNR) Signal-to-noise ratio (imaging) Signal to noise plus interference (SNIR) Signal-to-quantization-noise ratio (SQNR)

Related topics	List of noise topics Acoustics Colors of noise Interference (communication) Noise generator Radio noise source Spectrum analyzer Thermal radiation

Photography

Outline

Terminology	35 mm equivalent focal length Angle of view Aperture Black and white Chromatic aberration Circle of confusion Color balance Color temperature Depth of field Depth of focus Exposure Exposure compensation Exposure value F-number Film format Film speed Focal length Guide number Hyperfocal distance Metering mode Perspective distortion Photograph Photographic printing Photographic processes Reciprocity Red-eye effect Science of photography Shutter speed Sync Zone System

Genres	Abstract Aerial Architectural Conceptual Conservation Cloudscape Documentary Ethnographic Erotic Fashion Fine-art Fire Forensic Glamour High-speed Landscape Lomography Nature Nude Photojournalism Pornography Portrait Post-mortem Selfie Social documentary Sports Still life Stock Street Vernacular Underwater Wedding Wildlife

Techniques	Afocal Bokeh Brenizer Contre-jour Cyanotype ETTR Fill flash Fireworks Harris shutter HDRI High-speed Holography Infrared Kirlian Kite aerial Long-exposure Macro Mordançage Multiple exposure Night Panning Panoramic Photogram Print toning Redscale Rephotography Rollout Scanography Sabatier effect Stereoscopy Stopping down Strip Slit-scan Sun printing Tilt–shift Miniature faking Time-lapse Ultraviolet Vignetting Xerography

Composition	Diagonal method Framing Headroom Lead room Rule of thirds Simplicity

Equipment	Camera light-field field pinhole press rangefinder SLR still TLR toy view Darkroom enlarger safelight Film base format holder stock Filter Flash beauty dish cucoloris gobo hood hot shoe monolight Reflector snoot Softbox Lens Wide-angle lens Zoom lens Telephoto lens Manufacturers Monopod Movie projector Slide projector Tripod head Zone plate

History	Timeline of photography technology Analog photography Autochrome Lumière Box camera Calotype Camera obscura Daguerreotype Dufaycolor Heliography Painted photography backdrops Photography and the law Glass plate Visual arts

Digital photography	Digital camera D-SLR comparison MILC camera back Digiscoping Digital versus film photography Film scanner Image sensor CMOS APS CCD Three-CCD camera Foveon X3 sensor Image sharing Pixel

Color photography	Color Print film Reversal film Color management color space primary color CMYK color model RGB color model

Photographic processing	Bleach bypass C-41 process Cross processing Developer Digital image processing Dye coupler E-6 process Fixer Gelatin silver process Gum printing K-14 process Print permanence Push processing Stop bath

Lists	Most expensive photographs Photographers Norwegian Polish street women

Category Portal

This article is issued from Wikipedia - version of the 3/31/2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.