search

melix / astro4j

Astronomy libraries for Java
Apache License 2.0
26 stars 4 forks source link

Bad banding compared to INIT #345

Closed smart-underworld closed 5 days ago

smart-underworld commented 1 week ago

With the same ser file, I get bad banding of the continuum, which may be the source of the bad Doppler result Screenshot 2024-07-07 110609 Screenshot 2024-07-07 110632

melix commented 1 week ago

The cause might be the same as #344 , with incorrect limb detection, which would cause an invalid polynomial.

melix commented 1 week ago

It appears that the banding issue is unrelated. It's indeed very present on your image and seems to be interference. What was the setup you used?

smart-underworld commented 1 week ago

Since v2.6.0 won't generate the doppler images for you to debug, I used v2.5.3, with these settings:

config.json Screenshot 2024-07-08 151157 Screenshot 2024-07-08 151212 Screenshot 2024-07-08 151230 Screenshot 2024-07-08 151241

smart-underworld commented 1 week ago

I actually used a different scope from above. I updated the settings, but got the same result:

Screenshot 2024-07-08 151729

melix commented 1 week ago

Sorry I wasn't clear enough. It seems that the images you are producing have strong interference fringes. A common cause for this is a polarized ND filter. It is great that INTI can correct this, but as of now JSol'Ex will not. I have tried different ways to fix it, but none is really satisfactory, since while it would fix your image, it will make other images which don't have the problem look "flat", or, worse, introduce more artifacts.

Regarding the doppler images not being generated, you can try the 2.6.1 development version which should already fix this.

smart-underworld commented 1 week ago

Strange. I don't use any filters. Just a Herschel solar wedge.

On Mon, Jul 8, 2024, 11:58 PM Cédric Champeau @.***> wrote:

Sorry I wasn't clear enough. It seems that the images you are producing have strong interference fringes. A common cause for this is a polarized ND filter. It is great that INTI can correct this, but as of now JSol'Ex will not. I have tried different ways to fix it, but none is really satisfactory, since while it would fix your image, it will make other images which don't have the problem look "flat", or, worse, introduce more artifacts.

Regarding the doppler images not being generated, you can try the 2.6.1 development version which should already fix this.

— Reply to this email directly, view it on GitHub https://github.com/melix/astro4j/issues/345#issuecomment-2216746442, or unsubscribe https://github.com/notifications/unsubscribe-auth/BGFJTATVPFF2FMGFGDDBRNTZLOCY3AVCNFSM6AAAAABKPR2E7SVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMJWG42DMNBUGI . You are receiving this because you authored the thread.Message ID: @.***>

melix commented 1 week ago

Could it be that the Herschel wedge has a filter in?

smart-underworld commented 1 week ago

I removed the inner ND filter. Just glass reflection now.

On Tue, Jul 9, 2024, 6:25 AM Cédric Champeau @.***> wrote:

Could it be that the Herschel wedge has a filter in?

— Reply to this email directly, view it on GitHub https://github.com/melix/astro4j/issues/345#issuecomment-2217740167, or unsubscribe https://github.com/notifications/unsubscribe-auth/BGFJTAQN74UYEDBEGTGHWGLZLPQFRAVCNFSM6AAAAABKPR2E7SVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMJXG42DAMJWG4 . You are receiving this because you authored the thread.Message ID: @.***>

melix commented 1 week ago

Right. I don't know what could cause this, you'd have to talk to specialists like Christian Buil. I'll see if I can find a way to handle it, but at first glance seems fairly complicated.