3 ms·

I took one frame, frame 384 to be precise, where the system has locked onto the object and tracks it. The frame is 428 pixels wide and high. The AN/ASQ-228 ATFL

I took one frame, frame 384 to be precise, where the system has locked onto the object and tracks it. The frame is 428 pixels wide and high. The AN/ASQ-228 ATFLIR has three selectable fields of view and during the entire video it is set to NAR with zoom 1.0 wich means 0.7°. Dividing this by the frame dimension gives us the angular size of one pixel as 0.0016355°. The system reports the distance to the object as 4.4 Nm or equivalently 8148.8 m. Here one could argue that the system may report an incorrect distance but for now I will assume that value is correct. Multiplying the tangent of the angular pixel size with this distance to the object gives us the pixel size at this distance as 0.233 m. Finally we see that the locked objects covers an area 6 pixel wide and 4 pixels high which multiplied with the pixel size at the object distance gives us the camera facing object size as 1.395 m wide and 0.930 m high. width = tan(0.7° / 428 px) * 4.4 Nm * 6 px height = tan(0.7° / 428 px) * 4.4 Nm * 4 px I am not addressing your concerns now because they are not relevant for determining the size of the object. Any objections so far?

pmarreck 5y ago
Where are you getting the numbers from such as the equivalency of zoom 1.0 meaning 0.7 degrees? There's a lot of math here and although I don't initially disagree with it, any error at any point could result in a wildly different final estimate
- danbruc 5y ago
  The field of view is from the manufacturer [1], there are a some specifications at the very end. That the three options are indicated by WFOV (Wide FOV), MFOV (Medium FOV) and NAR (Narrow FOV) I first found in a flight simulator documentation [2] and it matches what you actually see in videos. The zoom is on top of the selected field of view, with a zoom of 2.0 the field of view would shrink to 0.35° in NAR. There is one thing that I noticed in the flight simulator documentation, the field of views are given as 6°, 3°, and 1.5° while the manufacturer states 6°, 2.6°, and 0.7°. That seems a bit odd as they are seemingly trying to make a very realistic simulation and I don't understand why they would deviate from the real values here. In the end I decided to just stick with the value Raytheon gives without digging too much into this difference. The frame and object sizes in pixel are certainly correct, everyone can check this. The distance is also easy to read in the video and therefore certainly correct unless you want to argue that the distance measurement didn't work properly and reported a wrong value. Finally the field of view should also be certainly correct, only that the flight simulator documentation gives different values causes me a tiny bit of doubt. Further questions about the size? If not, what's next? Altitude, temperature, speed, wings, ...? Or does the size alone convince you that it is just a bird? [1] https://web.archive.org/web/20091211103559/http://www.raytheon.com:80/businesses/rtnwcm/groups/sas/documents/content/rtn_sas_ds_atflir.pdf https://web.archive.org/web/20091211103559/http://www.raythe... [2] https://forums.vrsimulations.com/support/index.php/A/G_Advanced_Targeting_FLIR_(ATFLIR) https://forums.vrsimulations.com/support/index.php/A/G_Advan...
  - pmarreck 5y ago
    There was apparently (based on the audio recorded) a fleet of them (although this neither proves nor disproves the bird hypothesis). Could also be drones or missiles of some sort. But they're also claiming that this occurs on a daily basis, which supports the bird hypothesis I guess. The claims of physics-defying direction changes and speeds does not, though.
    - danbruc 5y ago
      Then speed it is? This will take me a few days, I started with size because it was quick and easy, but for the speed I will have to extract information from all or at least many frames instead of just looking at one.

pmarreck 5y ago

Where are you getting the numbers from such as the equivalency of zoom 1.0 meaning 0.7 degrees? There's a lot of math here and although I don't initially disagree with it, any error at any point could result in a wildly different final estimate

danbruc 5y ago
The field of view is from the manufacturer [1], there are a some specifications at the very end. That the three options are indicated by WFOV (Wide FOV), MFOV (Medium FOV) and NAR (Narrow FOV) I first found in a flight simulator documentation [2] and it matches what you actually see in videos. The zoom is on top of the selected field of view, with a zoom of 2.0 the field of view would shrink to 0.35° in NAR. There is one thing that I noticed in the flight simulator documentation, the field of views are given as 6°, 3°, and 1.5° while the manufacturer states 6°, 2.6°, and 0.7°. That seems a bit odd as they are seemingly trying to make a very realistic simulation and I don't understand why they would deviate from the real values here. In the end I decided to just stick with the value Raytheon gives without digging too much into this difference. The frame and object sizes in pixel are certainly correct, everyone can check this. The distance is also easy to read in the video and therefore certainly correct unless you want to argue that the distance measurement didn't work properly and reported a wrong value. Finally the field of view should also be certainly correct, only that the flight simulator documentation gives different values causes me a tiny bit of doubt. Further questions about the size? If not, what's next? Altitude, temperature, speed, wings, ...? Or does the size alone convince you that it is just a bird? [1] https://web.archive.org/web/20091211103559/http://www.raytheon.com:80/businesses/rtnwcm/groups/sas/documents/content/rtn_sas_ds_atflir.pdf https://web.archive.org/web/20091211103559/http://www.raythe... [2] https://forums.vrsimulations.com/support/index.php/A/G_Advanced_Targeting_FLIR_(ATFLIR) https://forums.vrsimulations.com/support/index.php/A/G_Advan...
- pmarreck 5y ago
  There was apparently (based on the audio recorded) a fleet of them (although this neither proves nor disproves the bird hypothesis). Could also be drones or missiles of some sort. But they're also claiming that this occurs on a daily basis, which supports the bird hypothesis I guess. The claims of physics-defying direction changes and speeds does not, though.
  - danbruc 5y ago
    Then speed it is? This will take me a few days, I started with size because it was quick and easy, but for the speed I will have to extract information from all or at least many frames instead of just looking at one.