https://siliconpr0n.org/wiki/doku.php?id=uvscada:gxs700
Think I might have to try this later! I am going up to an office tonight and they just happen to have a Dexis Sensor.
![]()
Looks easy enough with his instructions. What a cool project!
IDEA! Could we use this to create a calibration file??????? Other then creating a box with led walls to create a dark room without light.. and knowing what image format the dexis software wants the calibration files.. I mean.. it might work.
But as the author notes, playing around with X-Rays is no laughing matter, and you would want safety in mind for a project like that. I saw a home made box that worked great, it was wooden and had lead lining. A sensor would go inside and somehow the box was fitted for an xray head I think.
https://en.wikipedia.org/wiki/Flat-field_correction ?
I am going to keep thinking about this. Like.. "Why don't capture buttons work on cameras?" rofl.
Update: With some help from a friend, and looking at the github and the differences between resolution and actual resolution.. I can read a raw dexis calibration file, so I should be able to write it. I have forked that man's project in case he takes it down!
Take 4 bytes out of a candy bar and pay attention to the sign the little indian boy shows you.
Also, my sensei says it is best to take an average of 8 shots of flat and gain and create an image out of the averages. 16 is better.. ImageJ can do that as well. So the only thing left to do is alter the github project to compensate for the resolution difference, and figure out.. which part of the code does raw and tell it NOT to output to PNG, because then it wouldn't be raw. I could go into greater detail.....
For instance:
![]()
Think I might have to try this later! I am going up to an office tonight and they just happen to have a Dexis Sensor.
IDEA! Could we use this to create a calibration file??????? Other then creating a box with led walls to create a dark room without light.. and knowing what image format the dexis software wants the calibration files.. I mean.. it might work.
But as the author notes, playing around with X-Rays is no laughing matter, and you would want safety in mind for a project like that. I saw a home made box that worked great, it was wooden and had lead lining. A sensor would go inside and somehow the box was fitted for an xray head I think.
https://en.wikipedia.org/wiki/Flat-field_correction ?
I am going to keep thinking about this. Like.. "Why don't capture buttons work on cameras?" rofl.
Update: With some help from a friend, and looking at the github and the differences between resolution and actual resolution.. I can read a raw dexis calibration file, so I should be able to write it. I have forked that man's project in case he takes it down!
Take 4 bytes out of a candy bar and pay attention to the sign the little indian boy shows you.
Also, my sensei says it is best to take an average of 8 shots of flat and gain and create an image out of the averages. 16 is better.. ImageJ can do that as well. So the only thing left to do is alter the github project to compensate for the resolution difference, and figure out.. which part of the code does raw and tell it NOT to output to PNG, because then it wouldn't be raw. I could go into greater detail.....
For instance:
You would not want a 8 bit grayscale png file for the actual calibration file because calibration files need to contain the high bit depth fidelity in order to be applied to the raw 12-16bpp grayscale in the image - prior to converting to something for display on the monitor or window leveling.For your testing/learning you should try loading a calibration file into a software that can view 12-16bpp grayscale (that supports automatic min/max stretch or user/manual window leveling)
