Filters, explain the basics please

[SAW]

Hi,

I am using a 178MM for lunar and solar but I want to get a filter wheel to try some colour imaging on planets and even the moon although it's pretty colourless !

I already own a uv/ir and looking to get these https://www.firstlightoptics.com/baader-filters/baader-lrgbc-ccd-filter-set.html

I know I need to use R,G and B to get a colour image but do I use the uv/ir on it's own or in front of a colour filter ?

When, what or why would I use a uv/ir for on it's own ?

When would I use a clear filter ?

Thanks.

 

[geordie85]

I'd suggest getting rgb filters that already have ir/uv cut in them 

[kens]

The UV/IR cut filter is commonly known s the L or Luminance filter. The principle of LRGB imaging is that human vision perceives fine detail in luminance rather than in chroma (colour). So you combine an image taken with a luminance filter (on its own), with attention paid to capturing the fine detail, with RGB images which need not be as detailed. The result is a colour image with fine detail.

I've no idea if it is a technique used for planetary imaging but it is used for DSOs.

Also not sure of the use for clear filters other than to keep the sensor clean when its not in use.

[symmetal]

As well as visible light (R, G and B), the camera is also sensitive to UV and IR, wavelenths outside the visible spectrum. A simple convex telescope objective lens will only bring a narrow band of wavelengths to a focus at the sensor. If you focus with a green filter changing to red or blue you would need to refocus and if you wanted to image in IR or UV you would have to refocus even more. Telescope manufactures go to great lengths using multiple lenses to bring as wide a range of wavelengths (colours) into focus at the same point as they can. Triplet lenses can bring most of red, green and blue to the same focal point but IR and UV will still be out of focus.

With no filter the image from the camera will have sharp R, G and B wavelengths (if using a triplet) which is your luminance image, with out of focus UV and IR wavelengths superimposed on top so you'll get a smeared image. Using the UV/IR cut filter (L filter) you stop the out of focus UV and IR reaching the sensor.

When using an R, G or B filter it will block wavelengths outside that specific colour, so UV and IR will also be blocked so there is no need to use an UV/IR cut filter in tandem with a colour filter.

Using filters will alter the focal point of the scope due to the light travelling through a different medium (glass instead of air) and some LRGB filter sets are sold as being par-focal so that you don't need to refocus when changing filters. If you don't want to filter anything you use the par-focal 'clear' filter to avoid having to refocus. In reality many claimed par-focal filters are not fully par-focal, so re-focusing is still necessary.

Also some cheap RGB filters do exhibit an increase in transmission at UV and IR wavelengths so there is a case in using an IR/UV cut filter in tandem in this instance. The Baader RGB filters are good and have no UV/IR transmission.

Hope this helps SAW. ?

Alan

[SAW]

Thank you Alan ?

So if I had a RGB set plus UV/IR and IR Pass filters and wanted to image Jupiter with my mono cam I would use each filter individually for example 60s on each filter or 1000frames and then stack all avi files to great a single colour one ? I could then stack that in AutoStakkert for my final image ?

[symmetal]

2 hours ago, SAW said:

So if I had a RGB set plus UV/IR and IR Pass filters and wanted to image Jupiter with my mono cam I would use each filter individually for example 60s on each filter or 1000frames and then stack all avi files to great a single colour one ? I could then stack that in AutoStakkert for my final image ?

You would stack the videos separately in Autostakkert so you end up with 3 separate tiff files for R, G & B or 4 tiff files if you have an UV/IR video or IR pass video as well.

For planets if you have captured R, G and B you won't gain much with an UV/IR (L) as well as it will be significantly less sharp. Seeing and particularly atmospheric dispersion will spread the 'L' image around more than the individual colours. Using a one-shot colour camera on planets the RGB image you get after stacking will have the three colours mis-registered due to atmospheric dispersion. You can then register them in Photoshop etc. However, your 'L' image will have the mis-registered RGB components combined already. Using an 'IR pass' video and treating it as an L image can be worth trying, but I found it's much too dark at the exposures used for RGB and using much longer exposure times means you don't really beat the 'seeing' distortion. If you have a large aperture scope it would be useful. Using IR Pass is very good for the moon as it's much brighter and the IR end of the spectrum is less affected by 'seeing'.

So initially just take the three R, G, and B videos. Having obtained your three stacked mono tiff files from Autostakkert you can combine them into a colour image in Photoshop or various other programs and then do your enhancements, registering, de-convolution, wavelets etc.

Here's a useful web site giving the steps involved for RGB planetary imaging.

Good luck. ?

Alan

 

[SAW]

I'm slowly learning this LOL

For planetary then I would just be using R,G and B filters to give me my final image, the Baader ones all have IR coating.

For DSO then I would be also needing a Luminance as well ? Is luminance the UV/IR or clear filter ?

What exactly is the clear one used for or is that the luminance one ?

Can I image with my 178MM with no filter at all or should there always be something in front to protect the sensor ?

 

[symmetal]

Hi DAZ,

(1)  Yes that's a good starting method, and yes, the Baader RGB filters do block the IR and UV.

(2)  As DSOs are much fainter than planets, then capturing Luminance is very beneficial as you can capture the image with less exposure time than RGB. LRGB is usually the best method for DSOs. As exposures are much longer than when planetary imaging the effects of Atmospheric Dispersion on the Luminance compared to the RGB images is less important as the effects of seeing will be the limiting factor on resolution. If you're imaging DSOs near to the horizon however the Atmospheric Dispersion will be more prominent so not using L, and just imaging RGB in that instance could be considered. And, yes the L filter is the UV/IR cut filter.

(3) I did mention what the 'clear' filter was for a couple of posts ago though it's easily missed. ? It's just to avoid having to re-focus if you want to image without any light filtration, assuming that all your filters are par-focal (the focal point doesn't change when you change filters.). Many filter sets claimed to be par-focal aren't fully par focal, so refocusing may still be required. It's rarely used so if you don't already have one don't bother getting one. You could use it to 'protect' the sensor but using the L for this is more useful as you'd spend less time changing filters.

(4) You can, but the camera is sensitive to UV and IR so these will form out of focus images on top of your 'focused' image and so you end up with a 'smeary' images. I would always have the L (UV/IR cut) in place if you aren't doing RGB and want a luminance image. The IR-pass is a special case where you only want to record the IR wavelengths so you would use this on its own. UV, G, B and most of R are blocked with this filter.

Alan

[SAW]

Thanks Alan ? 

I've just got 5 position manual wheel for now so I'll put the uv/ir, and rgb filters in it and mayne an ir pass.

[symmetal]

Just to show how useful the L filter is here are two Autostakkert stacked images of Jupiter one with and one without an L filter using an ASI 224MC one-shot colour camera. No other enhancements have been done.

With L filter. Atmospheric dispersion slightly noticeable with blue fringe at the top-right and red fringe at the bottom-left.

Without L filter. Out of focus UV is added to the blue and out of focus IR is added to the red. Atmospheric dispersion effects are more evident as the IR and UV are 'bent' more than the R and B leading to vertical mis-registration and severe colour fringing and blurring. The moon is more oval than in the image with the L filter.

Alan

[SAW]

That's with a uv/ir filter ?

The first image is much cleaner/sharper looking.

[symmetal]

Yes, the UV/IR cut filter is commonly called the L filter. ?

Alan

[symmetal]

Here's the top image with the L filter after processing. (Astra Image De-convolution and sharpening plugins for Photoshop.)

Alan

[SAW]

Which scope is that with ? I've got no where near that with my Mak150 and a OSC, hoping I can better with the filters.

[symmetal]

It's a WO FLT98 and a 4x Powermate. Not an ideal planetary scope. At an effective 2472mm focal length and f25 it's the highest resolution I can get with the 98mm aperture. Your Mak 150 should be able to get a higher resolution due to its larger aperture. If you use a 2x barlow or Powermate you have a focal length of 3600mm at f24 which will give you a larger image than mine with a similar sampling scale compared to the Dawes limit (maximum resolving capability in arcseconds) of your scope.

Alan

 

[SAW]

What can I use to combine my separate rgb images in to one single colour image ?

 

[symmetal]

I'd use Photoshop, though other Astronomy programs like Startools, Nebulosity, AstroArt will do it although none are free programs. I believe ImageJ or it's Astronomy related version AstroImageJ will do it. I imagine Gimp would too. Someone who has used these free programs will be able to confirm, or suggest other options. ?

Alan

[SAW]

Hi,

Got to try out the filters last night on Jupiter ? I wondering if my problem of grainy images is me setting the image size to 600 x 600 with my 178MM ? I normally leave it at the default setting of 3096 x 2080 and get Jupiter into the centre of view and then adjust it to 600 x 600 which I find then it's quite grainy, I might be completely wrong here and not understanding adjusting the resolution though ! But it was just a though that maybe I should just try imaging at 3096 x 2080 ?

[symmetal]

Changing the captured image size won't have any affect on the image quality assuming you have the same exposure and gain settings. It just affects the number of frames you can capture per second, and the file size of your captured video. At full resolution I can only manage 30fps with a USB3 connection. If I reduce the image size to around 600 x 600 I can capture 150fps and at 400 x 400 it goes up to 200fps. Your maximum exposure time is determined by 1/FPS in mS, so at 200fps 5mS is your maximum exposure, while at 30fps your maximum exposure is 33mS. At 33mS exposure you can set the gain to zero and have a bright planet but the exposure will probably be too long to 'freeze' the 'seeing' so the images may be more blurred.

When imaging the moon I stick to full resolution, unity gain, with a short exposure, and record a longer video to make up for the drop in fps. For imaging planets at full frame you're capturing and downloading all that black background you don't need which makes your file sizes very large for no change in image quality.

I always capture several videos at different exposure and gain settings at each session to see what gives the best quality. At times of good 'seeing' you can give a longer exposure with less gain.

Image 'grainyness' should be improved by capturing and stacking more frames though you have to be aware of Jupiter's fast rotation so at your image scale 4 or 5 minutes should be your limit.

Alan

[Grierson]

On 31/05/2018 at 20:50, SAW said:

What can I use to combine my separate rgb images in to one single colour image ?

 

If you have Photoshop, ProDigital’s Astronomy Tools Actions are well worth the c. £18 IMVHO. For that you get over 30 Actions including building an image from RGB channels.