Canon Line Offset Noise

Started Oct 20, 2009 | Discussions thread
kolas Regular Member • Posts: 463
Re: It should work IMO

JimH wrote:

kolas wrote:

JimH wrote:

The question I've got is this:

Could a person come up with a good average value for these offsets for a given copy of the camera such that they'd be useful for other frames taken by the same camera?

I work in medical imaging field and exactly this is being done on digital X-ray detectors. A series of dark field and flat field (evenly exposed) images is taken to calibrate the detector for optimal results.

This seems like it would be extremely necessary for such systems. And it's got to make a huge difference given how the detection must be performed there.

Extremely low light, in photographic terms.

The medical "RAW converters" have been capable of correcting for uneven gains, barrell distortion and all kinds of artifacts and lens abberations quite some years ago. So the technology already exists and is developed to a great level of performance. I have seen some uncorrected images and could barely believe my eyes..

I'll bet they are pretty awful with no correction. With all of those separate detectors and the potential for geometric issues, I suppose that corrections are required before you get much of anything useful.

FYI most of the systems actually capture light, not X-rays, so the technology is very, very similar. One type of detectors is even based on optics and little CCD detectors, basically DSLR-size or MF-size sensors without Bayer RGB mask.

I remember looking into how the digital medical X-Ray detectors were implemented a while back. I do a lot of work with radiation detection equipment for laboratory and field use. It looked to me like arrays of scintillators coupled to light detector arrays was the standard way of doing this.

The so-called flat panels utilize a scintillator layer on TFT light detector matrix, CCD-based system have a separate scintillator screen with a WA lens pointing at it and a CCD detector. As the lens is basically a fisheye, there is a lot of geometric distortion that is automatically corrected by SW so that the output is a rectilinear image.

The same kinds of problems are present for both fields. Mainly, we want very high sensitivity so that low levels can be detected. After all, the medical folks don't want to roast the patients with any higher dose than necessary, and those of us doing environmental, prospecting, and research applications just want ever lower detection limits.

So you are familiar with the Radcal 1800ccm ionization chamber, right?

Btw. that is my dream, a dedicated BW DSLR, w/o the color filter array, with full luminance information in every pixel and ISO sensitivity shooting through the roof.

It's funny, but that's also been a dream of mine. Even though it seems like such a camera would be very cheap to build, I suppose that its limited appeal would still make it fairly expensive just due to the low demand.

That's a pity. Black and white is so wonderful. Imagine an 18 megapixel sensor with no CFA. Just a true 18 MP monochrome image with higher ISO sensitivity and a really weak (or no) AA filter. Maybe we're among the few who would like to see such a thing, but I'll bet there are at least a few others out there.

But I'd hate to have to pay a huge premium to get it.

DIY Basically all necessary parts are available off-the-shelf, they just won't fit into a SLR-sized body IMO.

-- hide signature --

Jim H.

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