J-20 5th Gen Fighter Thread VI

[vesicles]

I hope it works, it is the first time I post an image.

Please, Log in or Register to view URLs content!

The two sheets of paper in the image are of the same size, as you can see, the sheet of paper on bottom (position of the camera) is longer than the sheet of paper on top. Imagine now the bottom one is the J-20 and the top one is the J-16.

Of course the differerence of length is related to height of the position of the camera.

As you can see it in the picture of the J-20s and J-16s, the discrepance of how much the vertical stabilizer are showing backward, I think the difference is not neglectable, so therefore the J-16 looks shorter than it is.

The images we got were satellite images? When the satellite is so far above the planes on the ground, would the distortion still be significant?

 

[vesicles]

Can't we compare sizes knowing the diameter of the engines?
We know what kind of engines both planes uses, let's use the known diameter of them to made our calculations.

Good idea!

 

[latenlazy]

Can't we compare sizes knowing the diameter of the engines?
We know what kind of engines both planes uses, let's use the known diameter of them to made our calculations.

The smaller the measuring stick the greater the error. Besides, we don’t actually know the diameters of each engine at the nozzle to my knowledge, *and* there’s a clear widthwise perspective distortion too.

 

[jobjed]

The images we got were satellite images? When the satellite is so far above the planes on the ground, would the distortion still be significant?

Of course not!

With satellite, you can expect a non-classified resolution of 30cm/pixel at best. The 2048x1428 picture we got had a resolution of ~4cm/pixel. I'm not sure if the best military satellites have that good of a resolution, never mind civilian ones. It's obvious the picture we got came from drone photography.

 

[Hyperwarp]

Can't we compare sizes knowing the diameter of the engines?
We know what kind of engines both planes uses, let's use the known diameter of them to made our calculations.

I will give that a try. But I have to assume the AL-31 variant in the J-20 have the same diameter as the WS-10 variant in the J-16

 

[latenlazy]

I will give that a try. But I have to assume the AL-31 variant in the J-20 have the same diameter as the WS-10 variant in the J-16

They don’t. That’s part of the problem.

 

[Hyperwarp]

Ok, I must agree the J-16 has to be dropped from the measurement. Simple way is to look at the shape of the nozzles. J-20 nozzle looks nice and rectangular while the J-16 nozzles appear cylindrical. Cylinder viewed from above will appear rectangular but this does not apply to the J-16 nozzles. It is simply too far away and reinforces jobjed point.

So herer is a preview. What would be the nozzle diameter of the AL-31 series? I only have the intel diameter.

 

[kurutoga]

If you measure these blocks, they are all 88x73 pixels +/-1. So the degree of lens distortion can be ignored. Perspective distortion is also minimal. The reason for that is this is a drone picture. Drones usually have a wide lens. So the picture is very likely cropped so we are seeing only the center "good" area of this picture.

Normally lens distortion is non linear, you have to know the exact lens being used, and the focus length to make the correction.

Consider the circumstances this picture EXTREMELY UNLIKELY to be taken from a fixed wing aircraft doing fly-by, or a heli, because doing that is too expensive and may introduce motion blur.

 

[latenlazy]

If you measure these blocks, they are all 88x73 pixels +/-1. So the degree of lens distortion can be ignored. Perspective distortion is also minimal. The reason for that is this is a drone picture. Drones usually have a wide lens. So the picture is very likely cropped so we are seeing only the center "good" area of this picture.

Normally lens distortion is non linear, you have to know the exact lens being used, and the focus length to make the correction.

Consider the circumstances this picture EXTREMELY UNLIKELY to be taken from a fixed wing aircraft doing fly-by, or a heli, because doing that is too expensive and may introduce motion blur.

Except the planes are taller and closer to the camera than the ground, which is where perspective distortion would kick in.

 

[Quickie]

If you measure these blocks, they are all 88x73 pixels +/-1. So the degree of lens distortion can be ignored. Perspective distortion is also minimal. The reason for that is this is a drone picture. Drones usually have a wide lens. So the picture is very likely cropped so we are seeing only the center "good" area of this picture.

Normally lens distortion is non linear, you have to know the exact lens being used, and the focus length to make the correction.

Consider the circumstances this picture EXTREMELY UNLIKELY to be taken from a fixed wing aircraft doing fly-by, or a heli, because doing that is too expensive and may introduce motion blur.

Looking at the amount of space the vertical stabilizers overlapped the tail boom compared to what would be the case if the camera is directly on top, the amount of perspective effect on the apparent length would actually be quite a bit.

Imo, the fact that the blocks are still about the same size means that the picture was software manipulated to reduce the effect of wide angle lens (for aesthetic reason), especially since lens surface distortion would not explain the perspective effect of the VT and the tail boom mentioned above.

One can just hope that the software perspective adjustment are reasonably accurate, which should be the case judging by the almost equal size of those blocks.