Do optical WB filters improve RAW color latititude?

Started Aug 26, 2012 | Discussions thread
alanr0 Senior Member • Posts: 2,153
Sensor CFA design, optical WB filters and RAW color latititude

Vitruvius wrote:

So now I am getting oposing massages. One says, and I thought, that sensors are designed to "daylight' WB. Your post seems like you know your stuff too and I don't doubt that it may be true. But now I am more confused. Thanks though.

If the sensor were designed so that red, green and blue channels saturated uniformly under 'daylight' illumination, then you would need to under-expose the green channel whenever the colour temperature shifted appreciably. The eye is most sensitive to green wavelengths (around 555 nm), so with less green light on the sensor, perceived image noise would be degraded.

In practice, red and blue channels are made less sensitive than green.
For example: (page 14).

This means that for a useful range of illumination conditions, exposure depends strongly on the green channel, where the eye is most sensitive. Colour temperature can shift (within limits) without either red or blue channels clipping, provided the colour-weighted exposure is unchanged.

Rather than aggressively optimise performance for a single colour temperature, we for aim good performance with a much wider range of lighting conditions.

As John says, using a blue filter to re-balance tungsten illumination can improve signal to noise and colour accuracy in the shadows, provided you increase exposure time rather than pushing the ISO. For very long exposures (seconds and longer) you should also consider sensor leakage noise, which depends on the sensor design and degrades as operating temperatures rises.

Alan Robinson

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow