PIPP - planetary imaging pre-processor software

[Russe]

I do however have a modded colour SPC880 as well. But have not managed to get an RGB image with the mono cam.

[tonyowen]

V1.7 of PIPP has been uploaded to the website This version has support for writing the processed frames out as a .avi file. It has also had quite a lot of work done under the hood and has lots of minor bug fixes.

Cheers,

Chris

Chris

Downloaded v1.7 ran it to get rgb split, resulted in three avi files but Registax v5.1.9.2 would not load the avi.

error message "failed to decompress avi frame"

attached are relevant pipp files

regards

Tony

dubc+d_pipp_log.txt

dubc+d_pipp_quality.txt

[cgarry]

Hi Tony,

I notice from the log file that most of the frames were discarded. What size are the generated AVI files? Also can you run PIPP on one of the generated AVI files with the -debugavi option and post the output? Something like:

pipp filename.avi -debugavi

I have tested what you did here without any problems, but I have not managed to reject a large amount of frames which I think could be where the bug lies.

Cheers,

Chris

[tonyowen]

Hi Tony,

What size are the generated AVI files?

Also can you run PIPP on one of the generated AVI files with the -debugavi option and post the output? Something like:

Cheers, Chris

b = 4.77GB, g=4.48GB r = 4.18GB

will do.

regards

Tony

[cgarry]

b = 4.77GB, g=4.48GB r = 4.18GB

will do.

regards

Tony

They are much larger than I would expect since your logfile said only 42 frames were output. The output from the -debugavi run may be large in that case and best posted in a zipped text file.

Running this:

pipp filename.avi -debugavi > debug.txt

Will capture the output into a text file ready for zipping.

Cheers,

Chris

[tonyowen]

Running this:

pipp filename.avi -debugavi > debug.txt

Chris

I've tried ..... filename.avi -debugavi > debug.txt

both with and without spaces either side of the greater than sign.

In both cases all I get is return to d:\pipptest (and a flashing cursor) nothing else.

However if I do .. filename.avi -debugavi I get a debug.txt file under d:\pipptest

I've attached the file

debug.txt

[cgarry]

Running this:

pipp filename.avi -debugavi > debug.txt

Chris

I've tried ..... filename.avi -debugavi > debug.txt

both with and without spaces either side of the greater than sign.

In both cases all I get is return to d:\pipptest (and a flashing cursor) nothing else.

However if I do .. filename.avi -debugavi I get a debug.txt file under d:\pipptest

I've attached the file

Cool. That is the output for the source AVI - could you please do the same for one of the PIPP generated AVIs?

Thanks,

Chris

[tonyowen]

Cool. That is the output for the source AVI - could you please do the same for one of the PIPP generated AVIs? Thanks, Chris

Help, I do not understand request

The run, if I understand correctly, would be:

pipp -q=1000 -ctr -qw -rgb -qmin=2 -qmax=3 -qinc=1 d:\temp\pipp_20130416_164420\b_dubc+d_pipp.avi -debug.avi

[given that the > debug.txt option did not appear to function]

The generated avis are r, b & g, and the number of frames is far less than 1000 originally optioned.

So how does the -rgb and -q options work?

regards

Tony

[cgarry]

Sorry Tony, I am not making myself clear. All I want you to do if run 'pipp -debugavi' on one of the AVI files that PIPP generated but Registax will not open. None of the other options are need. So the command would be:

pipp d:\temp\pipp_20130416_164420\b_dubc+d_pipp.avi -debug.avi > debug.txt

The only file that should be generated is 'debug.txt' and that will be in the current directory of the command prompt. This debug.txt should have information about why it is not valid.

>> So how does the -rgb and -q options work?

They should work exactly as you used them and I have not managed to devise a test that breaks them yet.

Cheers,

Chris

[tonyowen]

So the command would be:

pipp d:\temp\pipp_20120416_164420\b_dubc+d_pipp.avi -debug.avi > debug.txt

Cheers, Chris

Chris, I think I've done it!!

Tony

debug.txt

[cgarry]

Perfect, that should help me with the debug.

Thanks,

Chris

[cgarry]

Chris

Downloaded v1.7 ran it to get rgb split, resulted in three avi files but Registax v5.1.9.2 would not load the avi.

error message "failed to decompress avi frame"

attached are relevant pipp files

regards

Tony

Have you tried this more than once? I am wondering if this fails every time or just once in a while.

Cheers,

Chris

[tonyowen]

Have you tried this more than once? I am wondering if this fails every time or just once in a while.

Cheers, Chris

Will repeat and advise

Tony

[tonyowen]

Will repeat and adviseTony

Chris, ran same avi file with same options thru pipp

the resulting r avi is a very faint mobile worm

the resulting g avi is more pronounced but still spotty

the resulting b avi is the same only highly visible.

sometimes I get the error "server execution failed"

running under Registax 5 gives a static image as shown

regards

Tony

dubc+d_pipp_log.txt

dubc+d_pipp_quality.txt

[neil phillips]

Looking foward to your user freindly version Chris. As soon as thats finished, I will be trying this out. Im no good to you on the debugging at the moment, as ive not used command lines. Any idea if or when you will have one in the pipeline ? BTW how have you been getting on with the 14" i remember you saying there might have been a optical problem. Which surprised me because the last Mars shot you took looked quality Chris.

[cgarry]

Chris, ran same avi file with same options thru pipp

the resulting r avi is a very faint mobile worm

the resulting g avi is more pronounced but still spotty

the resulting b avi is the same only highly visible.

sometimes I get the error "server execution failed"

running under Registax 5 gives a static image as shown

regards

Tony

Hi Tony,

Thanks for doing that, it is interesting that valid AVI files were generated this time.

Can you check what the resulting frames are like if you add the -bmp option so that the frames are output as individual bitmap images instead of AVI video files.

I notice that nearly all of the frames are being rejected in these runs because the planet is too small (no -ms=10 option), I wonder if the strange frames are the only ones with a big enough object to be selected.

During the rainy periods at Kelling, I experimented with your Saturn and Mars AVI's and realised that PIPP's standard quality algorithm just does not work on your captures. This is a combination of the image scale being very small and the fact your red and green channels are both overexposed giving very little detail to calculate quality from. Anyway, I have added a new histogram based quality algorithm to PIPP that does a substantially better job of rejecting the smeared frames. I shall email you a test version so you can try it out with your data.

Cheers,

Chris

[neil phillips]

Looking foward to your user freindly version Chris. As soon as thats finished, I will be trying this out. Im no good to you on the debugging at the moment, as ive not used command lines. Any idea if or when you will have one in the pipeline ? BTW how have you been getting on with the 14" i remember you saying there might have been a optical problem. Which surprised me because the last Mars shot you took looked quality Chris.

Oh ok then

[cgarry]

Looking foward to your user freindly version Chris. As soon as thats finished, I will be trying this out. Im no good to you on the debugging at the moment, as ive not used command lines. Any idea if or when you will have one in the pipeline ? BTW how have you been getting on with the 14" i remember you saying there might have been a optical problem. Which surprised me because the last Mars shot you took looked quality Chris.

Hi Neil,

The current plan for the GUI version is to have it ready before the next Jupiter opposition. I have a good idea of what I want it to look like and it should be a big step up from the current command line version.

As for the C14, it is still an open question as to whether there is actually a problem or not. Initial star tests did seem to show an issue, but they were inconclusive due to tube currents I was having problems with at the time. The good news is later tests seem to show that there may not be any spherical aberration problems but I still think there could be an astigmatism issue. Another star test with better seeing, better collimation and a completely cool tube should settle that issue.

Cheers,

Chris

[neil phillips]

Hi Neil,

The current plan for the GUI version is to have it ready before the next Jupiter opposition. I have a good idea of what I want it to look like and it should be a big step up from the current command line version.

As for the C14, it is still an open question as to whether there is actually a problem or not. Initial star tests did seem to show an issue, but they were inconclusive due to tube currents I was having problems with at the time. The good news is later tests seem to show that there may not be any spherical aberration problems but I still think there could be an astigmatism issue. Another star test with better seeing, better collimation and a completely cool tube should settle that issue.

Cheers,

Chris

Cheers for the info

Thought i became invisable for a sec there

[cgarry]

Cheers for the info

Thought i became invisable for a sec there

Not quite. I wrote the reply to you but did not realise for some time that I had hit the preview button but not the post button. Doh!

[neil phillips]

Not quite. I wrote the reply to you but did not realise for some time that I had hit the preview button but not the post button. Doh!

Arh ok no worrys. ive done it myself

cheers again

[tonyowen]

Hi Tony,Anyway, I have added a new histogram based quality algorithm to PIPP that does a substantially better job of rejecting the smeared frames. I shall email you a test version so you can try it out with your data.

Hi Chris

I downloaded pipp_pre-1.8. I ran it with my dubc+d avi which is a combination of the mars03 & mars04 files on the dvd. I also ran saturn03, again the avi is on the dvd.

I ran some of the resulting pipp avi's through Registax 5.1.9.2 on auto up to the Do it all command

Eight tests run with varying results. All log files attached.

Test01 used -hs & -ms=10, resulted in three jpeg images attached.

Test02 used only -hs resulted in smeared images

Test03 -ms=10 (no -hs) results like that of test01 one jpeg attached

Test04 -rgb -hs results smeared

Test05 -hs zero frames resulting

Test06 -hs -ms=10 mainly smeared images

Test 07 -hs -ms=10 -mp=150 resulted in a very few frames but okay.

Test08 -hs -ms=10 -mp=5 first few frames smeared. Unticked those when running Regsitax, resulting jpeg file attached

I'm not certain if this will clarify or confuse the issues

regards

Tony

test01_pipp_log.txt

test02_pipp_log.txt

test03_pipp_log.txt

test04_pipp_log.txt

test05_pipp_log.txt

test06_pipp_log.txt

test07_pipp_log.txt

test08_pipp_log.txt

[cgarry]

Hi Tony,

It is useful to know that I have access to the frames in your dubc+d avi, thanks for that.

Looking at your results I see that they have all used the standard quality algorithm. You definitely need to add the '-qhisto' option to the command line to use the new histogram based quality algorithm.

Also, for the small image scale these captures have, you will always need to set a small minimum size for planet detection (for example: -ms=10) otherwise all your good frames will be discarded because the planet is not large enough.

Cheers,

Chris

[tonyowen]

Hi Tony,

Looking at your results I see that they have all used the standard quality algorithm. You definitely need to add the '-qhisto' option to the command line to use the new histogram based quality algorithm.

Chris, I though I had.

I used the -help option and from there thought it was the -hs function??

Please advise

regards

Tony

[cgarry]

Chris, I though I had.

I used the -help option and from there thought it was the -hs function??

Please advise

regards

Tony

Ah, I see your confusion. The -hs option will stretch the brightness of each colour in each frame to be the same range. I would not worry about using this yet, but it can help with getting consistent colours in the final images.

The -qhisto is the option you want, it simply selects the new quality algorithm instead of the standard one.

[edit]

A good command line for you would be something like:

pipp -q=1000 -qhisto -ctr -c=150 -ms=10 "dubc+d.avi"

[/edit]

Chris