SonotaCo.JP
SonotaCo Network Japan Forum
SonotaCo.JP Forum Index
homeTop Page  FAQFAQ   検索検索   メンバーリストメンバーリスト   ユーザーグループユーザーグループ   登録する登録する 
 プロフィールプロフィール   プライベートメッセージをチェックするプライベートメッセージをチェックする   ログインログイン 

UFOCapture images shift between "Live" & "

 
新しいトピックを投稿   トピックに返信    SonotaCo.JP Forum Index -> UFOCapture BBS in English
前のトピックを表示 :: 次のトピックを表示  
投稿者 メッセージ
minador



登録日: 2010.11.08
記事: 8
所在地: Vail (Tucson), Arizona, USA

記事日時: Fri Dec 13, 2013 6:32 pm    記事の件名: UFOCapture images shift between "Live" & " 引用付きで返信

Hello SonotaCo,

We are close to triangulating a meteorite fall, and are very excited!

But we found an unusual issue while performing our data reductions.

In our version of UFOCapture, we have tonight noticed a counter clockwise rotation in the images when we toggle between the "Live" panel and the "Replay" panel. The difference was estimated at 9 degrees. I attempted to create screen captures, but the live output does not show up when taking a "PrtScn".

My question:
Are the scintillation mask images still oriented correctly with the video (which we use to calculate Alt/Az with UFOAnalyzer)? In which case the shift is just a display issue and does not affect output quality.

Or does this shift cause a mismatch??

We have attempted to shift the azimuth results by 9deg, but are unsure if this matters.

A ~9 degree shift would cause large error - one camera station is about 70 miles away from the suspected fall site, and my station is about 30 miles away. From the far camera station, this is a 18Km difference.

I am not sure what version of UFOCapture I am using, but it is about 3 years old.

Thank you for your time and for the fun/useful software!!

Cheers,
Minador

_________________
What we don't grow or hunt for, we must MINE from the earth!
トップに戻る
ユーザーのプロフィールを表示  
SonotaCo
Site Admin


登録日: 2004.08.07
記事: 9982
所在地: 139.67E 35.65N

記事日時: Sat Dec 14, 2013 10:48 am    記事の件名: Re: UFOCapture images shift between "Live" & & 引用付きで返信

Hi,
Thank you for using UFO series.
Its a pleasure for me to hear that they are working in many countries.

minador wrote:
In our version of UFOCapture, we have tonight noticed a counter clockwise rotation in the images when we toggle between the "Live" panel and the "Replay" panel. The difference was estimated at 9 degrees.

Sorry, I could not understand your situation. What is 9 degrees?

If the objects in a video is rotating compared with a stored one, it can be ...
1. Camera was moved and the direction was changed physically.
Is it?

Or... Please let me know your observation style first.
What is your FOV size(is it allsky or fisheye)?
Where is the FOV center direction? north pole? zenith? or else?
if you show me some of your captured peak hold image, it will help understanding a lot.

minador wrote:
I attempted to create screen captures, but the live output does not show up when taking a "PrtScn".

How about taking a real picture of display by digital camera or smart phone. Picture below is taken by smart phone.
You may be able to read a UFOCapture version on the operation sheet.

minador wrote:
Are the scintillation mask images still oriented correctly with the video (which we use to calculate Alt/Az with UFOAnalyzer)?
In which case the shift is just a display issue and does not affect output quality.
Or does this shift cause a mismatch??


It depends on kind of shift. Generally, there are two cases.
1. There are enough reference star image on the video (that contains a meteor) and the scintillation mask position is matched to start catalog on UFOAnalyzerV2(UA2),
then the all measurement is based on the ref stars on the video, so the result is always correct regardless of the happenings in the capturing stage.
2. If there is no reference star on the clip(such as of cloudy night one) and the stored profile was used for measuring, then the direction error may happen if the camera direction was changed even a bit. (this can be compensated on UA2, if the error is known)



IMG_20131214_085756.jpg
 説明:
taken by nexus5
 ファイルサイズ:  135.02 KB
 閲覧数:  6270 回

IMG_20131214_085756.jpg


トップに戻る
ユーザーのプロフィールを表示   投稿者のウェブサイトに移動
minador



登録日: 2010.11.08
記事: 8
所在地: Vail (Tucson), Arizona, USA

記事日時: Sun Dec 15, 2013 6:50 pm    記事の件名: Re: UFOCapture images shift between "Live" & & 引用付きで返信

Ok, we have been doing field work, and did not have time to reply. We actually have 2 issues. I did not describe my partner's issue correctly.

Issue #1
My issue was in UFOCapture v2.22 and was caused by the camera moving. My camera is mounted on a pole. I have a fisheye lens pointed to the zenith. I am now working on a solid mount and will stop using the pole.

My situation is #1, so my data is good. I have plenty of stars and they are matched very well all across the full sky (attached is an image).

Issue #2
My partners problem is with UFOAnalyzer V 2.28. I thought he was talking about UFOCapture (as I investigated, I noticed Issue #1 within my project).

He has noticed that when advances from the inital view to Frame #1, the FOV becomes warped. We are afraid that this is causing an error. Screen captures are attached. The difference is subtle by looking at the images within the post, but if you place the files in a folder and toggle them using Windows Photo Viewer, you will notice shifting and warping of the FOV (sort of like a reverse blink comparator).

He is using an ~100 deg FOV lens with FOV centered to the Northwest.

Can you tell us what is causing issue #2?

Thank you again for your time and help.



Problem 2 bad.jpg
 説明:
Bad image
 ファイルサイズ:  269.59 KB
 閲覧数:  6224 回

Problem 2 bad.jpg



Problem 1 Ok.jpg
 説明:
Start "Good" image
 ファイルサイズ:  271.34 KB
 閲覧数:  6224 回

Problem 1 Ok.jpg



Final alignment_red.PNG
 説明:
Good fisheye alignment
 ファイルサイズ:  217.49 KB
 閲覧数:  6224 回

Final alignment_red.PNG



_________________
What we don't grow or hunt for, we must MINE from the earth!
トップに戻る
ユーザーのプロフィールを表示  
SonotaCo
Site Admin


登録日: 2004.08.07
記事: 9982
所在地: 139.67E 35.65N

記事日時: Mon Dec 16, 2013 11:20 am    記事の件名: Re: UFOCapture images shift between "Live" & & 引用付きで返信

Hi Minador

Thank you very much for clear and detail explanation. I think I understand the situation.
Attached image is the diff of good and bad.
It seems two different profiles(such as different "yx" value) were used to plot marks on still image and on the frame of movie. It is a mystery

I checked with UA2 V2.28, but it does not appear on my clips and I cannot imagine the reason at all.
It seems there is some special reason that I cannot think of.

Could you send the following files that cause this problem to me by some method?
M20131211_021053_SV1_Se.avi
M20131211_021053_SV1_SeP.jpg or bmp
M20131211_021053_SV1_SeM.bmp
M20131211_021053_SV1_SeA.XML
I will use these for debugging only.

My Email is
xxxxx@xxxxxxxxxxxxx(eraced)
But for the transfer of the avi file, please use some free file transfer service such as below, because avi is too big.
https://www.filemail.com/

Thank you.



problem_1_diff.jpg
 説明:
diff of good and bad ....
 ファイルサイズ:  384.49 KB
 閲覧数:  6201 回

problem_1_diff.jpg




最終編集者 SonotaCo [ Mon Dec 16, 2013 2:12 pm ], 編集回数 1 回
トップに戻る
ユーザーのプロフィールを表示   投稿者のウェブサイトに移動
minador



登録日: 2010.11.08
記事: 8
所在地: Vail (Tucson), Arizona, USA

記事日時: Mon Dec 16, 2013 1:12 pm    記事の件名: 引用付きで返信

Yes, very strange shifting. The combined image shows the issue very well.

I have emailed the files - there is no *P.bmp or .jpg. I generate a peak hold, but the other station is not set up to do that. I will request them to generate one.

I have sent an invitation for you to retrieve the avi by Dropbox.

_________________
What we don't grow or hunt for, we must MINE from the earth!
トップに戻る
ユーザーのプロフィールを表示  
SonotaCo
Site Admin


登録日: 2004.08.07
記事: 9982
所在地: 139.67E 35.65N

記事日時: Mon Dec 16, 2013 7:21 pm    記事の件名: Half of the mystery 引用付きで返信

Thanks to the files, I found the first cause of UA2's strange behavior.
The size of avi and the still images are different! and it cause the confusion of UA2.
The avi was 720x480, but the M.bmp and S.bmp was 640x480.

Usually the size of avi and bmps are always equal.
I cannot imagine how to create different size images, I have never seen such case.
Please let me know if you know how to create 720 avi and 640 bmps.
But anyway, it may be a problem of UFOCaptureV2 setting.
Please confirm it with the attention for the created media size.

As for the meteor that was contained in the clip.
I will send you the analysis result on my PC.
I expand the M.bmp to 720, and P.bmp was made by peakhold re-creation function of UA2 (on uty sheet).
Number of ref stars is 11, with 0.3 degree accuracy.
The trajectory was measured until the explosion, 2.25 sec.
This A.XML can be used for the profile of the system if the camera direction is tightly fixed.
And the M*.csv can be used for UFOOrbitV2.



M20131211_021053_SV1_Se_P.jpg
 説明:
re-created P.jpg and the measured trajectory
 ファイルサイズ:  168.62 KB
 閲覧数:  6184 回

M20131211_021053_SV1_Se_P.jpg


トップに戻る
ユーザーのプロフィールを表示   投稿者のウェブサイトに移動
特定期間内の記事を表示:   
新しいトピックを投稿   トピックに返信    SonotaCo.JP Forum Index -> UFOCapture BBS in English All times are GMT + 9 Hours
Page 1 of 1

 
移動先:  
新規投稿: 不可
返信投稿: 不可
記事編集: 不可
記事削除: 不可
投票参加: 不可
このフォーラムで添付ファイルを投稿 できません
このフォーラムでファイルをダウンロード できます


Powered by phpBB © 2001, 2005 phpBB Group
Copyright ©2004 SonotaCo Network. All Rights Reserved.