Fundamental question: What's a pixel?

Started Mar 7, 2009 | Discussions thread
Gyles Regular Member • Posts: 201
Re: #1 but not quite

Michael Fryd wrote:

There's a huge difference between a situation where scientists
understand the mechanisms and where they don't.

Scientists don't claim to understand the stock market. They claim to
have models that are useful in some situations, but they realize the
limitations of the models.

There are many situations where we have a very good understanding of
the situation, and we can accurately compute un-measured information,
based on related measurements.

For instance, a traditional mercury thermometer does not directly
measure temperature, it measures the volume of the mercury. A simple
calculation is performed, and we compute the unmeasured temperature
based on the measured volume.

Just because something isn't directly measured, does not mean that
the value determined is a 'guess.'

This is all true, and as others have said is a good analogy to the light measuring done by the sensor, but there's one other step. Each pixel need red, green and blue data, but only measures one of the three.

To push the thermometers analogy well beyond the point of usefulness say you have an office and want to know the temperature of every desk, but only have thermometers on one desk in three. You can use the measurements you have and some analysis, observation and maths to estimate the temperatures at the in-between desks. It's not a guess, but nor is it a measurement. It is probably good enough for most practical purposes, but can't account for (say) a localised cold draft or a patch of sun through a window on an unmeasured desk.

Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow