How do CCD pixel values scale with light, and what does gamma correction do?

Asked 7/5/2018

1 views

2 answers

0

I'm using a fluorescent microscope with a CCD camera to capture images. How do the recorded pixel values relate to the amount of light reaching each pixel? Is this scale the same for all CCDs, or does it vary between sensors? Also, what effect does gamma correction have on those pixel values?

Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0

Photography Stack Exchange contributor

8y ago

2 Answers

1

Digital imaging sensors are linear in their response to light. If you expose one to twice as much light, either by making the light twice as bright or by exposing for twice as long, the amount of voltage produced by each sensel will double until full well capacity is reached.

Different sensors will be more efficient or less efficient than other sensors, but they are all linear in their response. Simple amplification, that is, multiplying all measured voltages by the same amount of gain, is all that is needed to make a less efficient sensor output the same strength of signal as a more efficient sensor when both are exposed to the same amount of light.

Gamma correction is an operation by which the linear response of imaging sensors is converted to a logarithmic response that mimics the response of the human vision system. We are more sensitive to minor differences in brightness in a moderately lit scene than we are to differences in brightness of very bright or very dark scenes.

(Note that gamma correction in the processing of a digital image is not the same thing as what we mean when we say gamma correction with regard to the output of a video signal when it is send to a display device. Although the concepts are related, they are two different things done at different points in the processing pipeline between capturing an image with a digital sensor and viewing an image on an emitted light device.)

Originally by user15871. Source · Licensed CC BY-SA 4.0

user15871

8y ago

0

AI Answer

Generated from our catalog & community — verify before relying on it.

A CCD’s basic response is linear: if a pixel receives twice as much light, it produces about twice the signal, up to the point where it saturates (full well capacity). That means pixel values normally scale proportionally with light intensity and/or exposure time.

Different CCDs can vary in efficiency and sensitivity, so the same light may produce different signal strengths on different sensors. But the underlying response is still generally linear; differences are often handled with gain/amplification.

Gamma changes that relationship. Instead of keeping values proportional to light, gamma correction remaps the linear sensor data into a non-linear curve, typically to better match how human vision perceives brightness. In practice, this compresses some tonal ranges and expands others, so the stored/displayed pixel values no longer correspond linearly to the original light intensity.

So, for measurement and scientific imaging, you generally want the linear data. Gamma-corrected data is more appropriate for visual display than for quantitative analysis.

UniqueBot

AI

8y ago

Your Answer