Here's a little work I've done in order to try to better understand what it's all about here.
The intend of this work is to see if there's any detectable variation of luminosity in the "UFOs" and, if so, if it could apply to a logical
explanation, the flapping wings of birds, for example. (Or the regular luminosity pulsation of the UFOs lights...
I will firstly explain step by step the process I used, this way anybody should be able to verify my methodology (I could be wrong!) and to reproduce
1- Video capture
The original video max resolution is, on Youtube, of 720p HD. To upload such videos, I generally use Video
which allows you, without any line to wait.
After a few seconds or so, the process is done and a window pop up with multiple choices. Here, the best resolution is the first one (HD Quality, MP4,
Unfortunately (and we will see later why), the output file is now in MP4 (there's no choice from Video Grabber), but the following steps will allow
me to easily convert it.
2- Selection of the sequence
The full video makes 5'04" and have 9084 frames. For the studied sequence, I choose a part of it, in which there's no move from the cameraman and
no luminosity variations (such as those that can be seen at the end of the video when an object appears in the FOV and strongly reflect the IR
So I choose a 451 frames sequence, at the beginning, from frame 726 to frame 1177, with Virtual Dub
, a cool
video capture/processing software:
The other interest of this software (among many other things) is that it's able to re-save the video file using a full encoding process in .avi so,
that's what I've done with my selection.
3- Addition of a fixed white reference point in the sequence
Now that I have my .avi sequence, the next step is to study it in details with another wonderful software:
, dedicated to astronomical studies, but that can be used as well for UFOs studies, as
it use powerful analysis tools for both in the same way.
The next step is to compare the luminosity (or magnitude) variation of the object in time with a fixed reference (which luminosity don't vary) point
in the video, like a star for example.
However, at first, and it's clearly visible in the video, there's a strong atmospheric turbulence, which have for consequence to make the couple of
visible stars here to strongly twinkling. With IRIS, this unwanted effect could be theoretically fixed by using a special post-process, but in order
to do this, one needs some camera reference shoots, besides other things, that I do not have. (offset, dark, flat-field, etc.)
So, the trick here is to artificially put in the video this reference point, i-e a single white dot, in the same way (sort of...) we use in astronomy
a coronagraph in front of the sun to block out its light.
Windows Movie Maker, for example, can do that:
This white dot will not change its luminosity at all.
Last step here, before proceeding with IRIS, is to re-save the MP4 file created by Windows Movie Maker in .AVI with Virtual Dub.
4- Process with IRIS
The next step are:
- Open the .AVI file with IRIS
- Convert it in .FIT files, one file by frame
- Select one or several "UFOs", then select the "white dot" reference point.
- Select "automatic photometry" with up to five "UFOs" objects to track down:
- Open "verif.dat" file that sum up all the data in graph:
This graph represent the magnitude/luminosity variation (in thousandth of magnitude) of one of the "UFO" comparatively to magnitude/luminosity of
the reference point (that don't change at all).
Let's process now the same way with a visible star
in the whole sequence to see how the graph looks like:
5- What does that mean?
The graph for the star varies from -2080 to 2873, (if we exclude the very last frames - 424 to 451 - that returned errors due to a bad tracking) then
a total variation of 4953.
The graph for the UFO varies from -1563 to 10.000, (we can reasonably exclude -4710 as it is really a dark point), then a total variation of
The variation of the graph for the star is mainly due to the strong twinkling and this effect possibly apply as well for the "UFO" graph, as we
don't know the exact distance of these objects to the camera. The farthest the object is, the more the twinkling effect is visible.
...to be continued...