#1  
Old 18-09-2012, 09:24 AM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Photon Transfer Curve

Has anybody attempted to characterize their CCD camera by plotting a Photon Transfer Curve? I just got a new camera and I'm going to give it a go. If anybody has useful comments on the process or value of this exercise please let me know. I will post results here when I'm done...

Cheers,
Rick.
Reply With Quote
  #2  
Old 18-09-2012, 11:45 AM
rally
Registered User

rally is offline
 
Join Date: Sep 2007
Location: Australia
Posts: 896
Rick,

What camera are you talking about ?
Just wondering, if its not a cooled camera how you would control the CCD temperature ?
If it is an uncooled camera, how could it be relevant

As a method to compare CCDs for performance and quality its possibly very useful, as an exercise for your CCD experience in general maybe useful, as a method to characterise your system for some more specific purpose ?, I am not so sure.
What exactly are you hoping to get from the exercise that can be applied to your imaging ? - and improve it.
Especially if done in isolation

You have a calibrated integrating sphere to do this ?

I cant help but think that doing all the normal things and examining the S/N, image scale, well depth, spectral response, pixel and column defects etc etc and how that affects your images.
A couple of the semipros in the US have some good info and tests on their websites for doing that sort of characterisation. I would think that is more useful than a PTC.
I recall a piece of software written by some Europeans was available too

But then is this going to help you take better images ?
Understanding the maths and principles behind imaging noise, stacking, shot noise, darknoise, read noise, flats, bias etc etc and implementing best practice for each and every one of these problems, and all the others) is likely going to result in a greater benefit.
Not to mention eliminating all the other 'noise' in the rest of your systems

But without knowing what you are hoping for its probably a bit hard to comment really.

Rally
Reply With Quote
  #3  
Old 18-09-2012, 12:39 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Rally,

The camera is an Apogee Alta U16M with D9 body so it is very definitely cooled.

A lot of basic info can be determined from the PTC: full well depth, gain, read noise and Photo Response Non Uniformity. From what I've read (mainly a couple of presentations by Richard Crisp) the data from PTC analysis can be used to determine an appropriate operating temperature for the camera and to determine an optimal flat fielding protocol.

I don't have a calibrated integrating sphere but as far as I can tell all I need is a lightbox and a spreadsheet.

Cheers,
Rick.
Reply With Quote
  #4  
Old 18-09-2012, 01:10 PM
rally
Registered User

rally is offline
 
Join Date: Sep 2007
Location: Australia
Posts: 896
Rick,

The Apogee will do !! and Richard Crisp is one of the US semipros I was referring to !
If he says its good then I'd go with that ! He does that sort of stuff for a living.

I still think that a PTC is just one of the characterisations and just one of the noise reduction activities you could do and that even if you do this, you still need to rigourously do all the other things that are best practice.
No point identifying something that might produce a 1% improvement in FWHM or a S/N ratio and then ignore all the other many 'something elses' that each contribute 5% each (summed quadratically) to the noise in your system !
The question also becomes how much time have you got to attend to all of these and which ones are going to give you best bang for buck.

Look forward to the results and improvements.

Rally
Reply With Quote
  #5  
Old 18-09-2012, 10:49 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
I did an initial rough analysis and the results look sensible. Gain and read noise match the specs at 1.4 and around 9e- respectively. PRNU is 0.87% and full well is around 91Ke-. I need to do a few more data points and then try a Dark Transfer Curve and Flat Field Transfer Curve.

Cheers,
Rick.
Reply With Quote
  #6  
Old 19-09-2012, 07:30 AM
[1ponders]'s Avatar
[1ponders] (Paul)
Retired, damn no pension

[1ponders] is offline
 
Join Date: Nov 2004
Location: Obi Obi, Qld
Posts: 18,778
Hi Rick. I did it a few years ago for a Uni assignment (Comparing low end consumer grade CCDs and CMOS for use in scientific measurements). In AIP4Win there is a process described on how to measure the transfer function, linearity etc. I built the lightbox and screens for the process so if your interested let me know, your welcome to borrow them and the book to run through the process.
Reply With Quote
  #7  
Old 19-09-2012, 09:12 AM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Thanks, Paul. I have a copy of HAIP and I have done the basic test they describe on my SX camera. I was too lazy to build the test gear for the advanced version I'll go back and read it again and see if it will tell me anything I can't get from the PTC analysis. If so, I'd be interested to borrow the lightbox and screens some time.

Cheers,
Rick.
Reply With Quote
  #8  
Old 19-09-2012, 11:58 AM
[1ponders]'s Avatar
[1ponders] (Paul)
Retired, damn no pension

[1ponders] is offline
 
Join Date: Nov 2004
Location: Obi Obi, Qld
Posts: 18,778
Just give me a hoi when you need them Rick, they're sitting there gathering dust.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 01:42 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement