Black Calibration

From apertus wiki
Revision as of 07:56, 6 February 2016 by A1ex (talk | contribs) (Created page with "===What does this mean?=== Finding the sensor output value, per pixel, in the absence of any illumination. It covers: - dark frame subtraction - dark current compensation - ...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

1 What does this mean?

Finding the sensor output value, per pixel, in the absence of any illumination.

It covers: - dark frame subtraction - dark current compensation - using black reference columns (called "optical black" by other manufacturers) to find the black level and to correct some of the row noise

Note: black reference columns can be used to reduce row noise, that's why we include it here. That means, we will correct some (but not all) of the row noise as part of black calibration process.

2 Calibration methods

2.1 Dark frame subtraction

This is a basic technique: take a picture with the lens cap on. To make really sure no light is reaching the sensor, also cover the entire camera with something.

2.2 Dark current

2.2.1 Simple correction
2.2.2 Dark current nonuniformity correction

2.3 Black reference columns

This sensor has 8+8 columns that can be used for calibrating the black levels; they are also useful for reducing the dynamic row noise.

2.3.1 Black level
2.3.2 Row noise correction from black columns

3 Proposed calibration pipeline

Step 1: use black reference columns to find the black levels for odd and even rows Step 2: subtract dark offset and dark current Step 3: use variations in black reference columns to reduce row noise

4 Calibration procedure

4.1 Acquiring the images

4.2 Creating the reference frames

4.3 Using the reference frames to correct raw12 files