Apparent failure of ImageRegistration

[jamesRC]

Hello QuestionPeople,

My results from manual postprocessing of a set of CR2 raw images, indicate that ImageRegistration is failing to work. The integrated lights look like I would expect except that the star images are broken up into jagged lines made up of dots. They remind me of dithering patterns, fwiw. I tried re-registering the already registered images, and the star images are still not aligned. Everything about running the registration process works normally. What might I be running into here?

Thanks very much,
James

James

[jamesRC]

Hi,
More about image registration. I was able to  register my images of M51 using other software. It's crude because no calibration, and the "other" software produces FTS files which don't fit into a process that expects CR2 files

Anyway- below is an image of the registered image after integration. I see there is a fairly regular irregular pattern scattered about, it looks like a gritty smear odf something.
I'm wondering if the presence of these smears might be interfering with PI star registration? Is this possible?

James

[pfile]

they are hot pixels; it is possible that StarAlignment has interpreted them as stars. i am not sure if it can be done within BPP, but in StarAlignment you need to increase the hot pixel removal control and possible increase the noise scale control. one problem with the m51 area is that there are not a whole lot of stars, and so there are fewer stars for SA to latch on to. sometimes if the stars themselves are soft (meaning not very peaky) then you need to increase the "peak response" control as well, to allow for softer stars to be detected as stars.

rob

[jamesRC]

Thank you again, Rob.

Looks kinda like my attempt to rejuvenate my canon t3 may not have been worth it darnit.

James