Full search page →

DLCV 09C · Gain cannot create missing photons

AdvancedDLCV 09C · Low-Light Image EnhancementImage lab

Description

Simulate a dim photo with photon and read noise, then compare more light, more digital gain and a denoiser.

Try this: At 8 photons and gain 12 the picture is as bright as the scene, yet very noisy. Lower the gain to 1: the picture goes dark, but does the signal-to-noise ratio change? Now keep gain 12 and raise the photons to 100. What changes instead, and what happens to the highlights?

Predict first

You double the digital gain while nothing clips. What happens to the signal-to-noise ratio?

Hints

The chapter section that uses this lab:

Explanation

Answer the prediction first, or open the explanation when you are ready.

Show the explanation

Gain multiplies the signal and the noise by the same factor, so their ratio does not move; the picture only looks brighter and its noise looks larger. The ratio improves only when more photons are collected: signal grows with the photon count n while photon noise grows only with its square root.

Lab · DLCV 09C · Gain cannot create missing photons

Predict first

You double the digital gain while nothing clips. What happens to the signal-to-noise ratio?

Loading image…

The scene (what we want)
Sensor output after gain
After the enhancer

Try this: At 8 photons and gain 12 the picture is as bright as the scene, yet very noisy. Lower the gain to 1: the picture goes dark, but does the signal-to-noise ratio change? Now keep gain 12 and raise the photons to 100. What changes instead, and what happens to the highlights?

Camera test image, 256 × 256. Each pixel collects Poisson(photons × scene) electrons plus Gaussian read noise (σ = 2 electrons), then the output is clip(gain × electrons / 100). Noise fields are fixed, so sliders never re-roll them. SNR is the mean signal over the standard deviation of the noise, measured before clipping, in dB. The enhancers are a Gaussian blur (σ = 1.5) and a 3 × 3 median, stand-ins for a learned denoiser; they are not one. Large photon counts use a normal approximation of the Poisson distribution.

Method and image sources

Open in the Playground →