120 frame lunar mosaic

[Dave Smith]

Well finally I have managed to get a result from ICE (Microsoft Image Composite Editor)

To start from the beginning, processing 120 avi files takes a considerable time and I found it was useful to bring my old laptop back into use and had two computers going at the same time. This helped obviously to speed up the process but it also gave the brain more to think about so wasn't so boring.

As has been seen in my earlier posts, I had considerable problems in stacking the images in ICE due to variations in brightness of the images and in iMerge due to field rotation. Finally I tried dimming any frame that looked remotely bright and fed them all back into ice and got the following result (8% of full size)

At first sight this looked fine but then I noticed that it was very distorted. The top part is stretched vertically compared with the rest. So back into photoshop it went and I applied a transform to get it back to shape.

The final result is far from perfect but I am quite pleased to get a reasonable result. Below is a 25% version.

To see a 50% version see http://astrosnaps.co.uk/Moon/Moon-2009-10-09-final50.jpg

To see the full size version see http://astrosnaps.co.uk/Moon/Moon-2009-10-09-final100.jpg

I have questions about dealing with field rotation but I'll put them in a separate thread.

Thanks for looking.

Dave

[PaulG]

Don't whatever you do give up on this, it's looking really good.

You are making great progress and from what I can see it has been worth every effort.

I'm sure the more experienced SGL members will be able to help you fine tune your technique.

Paul

[johnh]

A mammoth project Dave has turned out brilliantly, I cannot get my head around 120 avi's to process it would take me months. Great dedication - encouraged me to have a go at a mosaic eventually.

Thanks for showing Dave. JohnH.

[resonator77]

A mammoth project Dave has turned out brilliantly, I cannot get my head around 120 avi's to process it would take me months. Great dedication - encouraged me to have a go at a mosaic eventually.

Dave

Brilliantly executed image:eek:

Can you give me some pointers as to what equipments were used for this image?

Thank you

Resonator77

[RobH]

You need a medal for the amount of sheer time and dedication you've put into this Dave

Great result at the end of it too, and I would imagine a very rewarding learning process too.

Cheers

Rob

[Beyond_Vision]

10 out of 10 for patience Dave I have been pulling my hair out doing a four pane DSO image and you have had to contend with 120. Well done and putting it all together

Regards

Kevin

[Dave Smith]

A mammoth project Dave has turned out brilliantly, I cannot get my head around 120 avi's to process it would take me months. Great dedication - encouraged me to have a go at a mosaic eventually.

Dave

Brilliantly executed image:eek:

Can you give me some pointers as to what equipments were used for this image?

Thank you

Resonator77

Thank you all for your encouraging comments. I forgot to say that this image (or set of images) was taken on the morning of 9th October, last Thursday from approx midnight until 3am when the clouds started to roll in.

Equipment details: 9.25" Celestron SCT with a DMK21AU04.AS Imaging Source camera at prime focus. The camera produces an avi file of around 2000 frames at the rate of 60 frames per second. Each avi is then sorted, stacked and processed in RegiStax5.

The 120 resulting images were thrown into ICE to stitch and finally a little tweaking in photoshop.

Dave

[MikeD]

Congrats Dave great detail showing and you got it all.

If I was you I would give it a go in imerge I did a 170 frame mosaic in 2005 and had problems with rotation I ended up having to stitch the image together in three different runs ( top middle and bottom) and then fit the three together in photoshop.

The hard work is done, stacking the 120 avi's.

You could try imerge in three runs and then stick them into Ice.

How tight is your polar alignment good polar alignment will reduce the rotation to a minimum, not much help now but worth a look before another attempt.

I was imaging the moon the same night but I am not sure if I will continue with the processing for a few reasons: I messed up the first run of five avi's by imaging too soon after clearing the dew off the secondary with a hair drier (turbulence still present) after catching the blue data for a DSO, the seeing not being the best and being restricted to 30 FPS due to a motherboard problem.

I'll post what I have so far in another thread.

Mike.

[neil phillips]

Thats a lot of work, but a good result

[resonator77]

Thank you all for your encouraging comments. I forgot to say that this image (or set of images) was taken on the morning of 9th October, last Thursday from approx midnight until 3am when the clouds started to roll in.

Equipment details: 9.25" Celestron SCT with a DMK21AU04.AS Imaging Source camera at prime focus. The camera produces an avi file of around 2000 frames at the rate of 60 frames per second. Each avi is then sorted, stacked and processed in RegiStax5.

The 120 resulting images were thrown into ICE to stitch and finally a little tweaking in photoshop.

Dave

Thanks for the info Dave,

why is it 120 avi ? is there a reason for this?

did you use video capture or single frames?

sorry for asking many questions but nobody mentions these for a newbie like myself:o

[Dave Smith]

Thanks again for the helpful comments.

Mike: I quite like your idea of mixing iMerge and ICE but probably for a future occasion. This time most of the Moon seems ok but there are definite problems at the limb near 7 o'clock. What I might try is put the entire image in photoshop, cut out the offending section(s) and repaste them either in PS or iMerge. The other important thing I will do is ensure that I keep the camera settings constant after any crash. I believe this was the main reason for having any problem with ICE. Next time I will also try a smaller phase (this one was 75%). I don't have any polar alignment as my 9.25" is fork mounted and I don't have a wedge, hence there is quite a lot of field rotation over the whole imaging session. (of the order of 20 degrees).

Resonator77: No problem, ask away - that's what the forum is about.

The camera takes a video sequence of just one part of the Moon. That sequence consists of around 2000 frames but all of the same view. The telescope does not move (relative to the Moon during that time). That video sequence in the form of an avi file is fed into RegiStax5. That program basically compares all 2000 images and puts them in order of merit compared with a "good" frame that you choose. It then stacks the best e.g. 300 to produce a superior image as the noise is averaged out. There is some further processing that sharpens the image and the final output is a single picture of one small section of the Moon. You then need this repeated enough time to cover the whole of the Moon with plenty of overlap between the images. I just happened to need 120 to do that. I didn't count them or decide in advance how many were needed but just started at the top right, took a sequence, moved the scope left ensuring some overlap and repeated until reaching the left edge and the moved down a bit again ensuring some overlap and basically zig-zagging down the Moon until it was all covered. The problem is the sticthing those 120 images together to produce the large image without any of the joins showing. I hope that helps.

Dave