ANALYSIS OF THE Aug. 17 2006 CHINESE UFO VIDEO by Bruce Maccabee

This video has some odd characteristics which appear to be "fingerprints of a hoax."


The image of the building goes from out of focus to in focus but the edges of the UFO image remain "fuzzy" (not sharp). Compare the relative "fuzziness" of the edges of the building and of the UFO in the following two pictures, the first at 2 seconds and the second at 6 seconds into the video.

2 sec

In the picture below the building image is much better focused but the UFO seems to have the same degree of defocus.

6 sec


The magnification or image size of the building changes with the "zoom out" (8 to 12 seconds) and then "zoom in" (13 to 17 seconds) by much more than the image size of the UFO changes. Consider the images below and the following table of data regarding the relative widths of the building image (see arrows in the illustration above) and the UFO image (width). Note that the left and right ends of the UFO image are "fuzzy" (not well focused) and this makes it difficult to determine the exact width of the image.

12 seconds 17 seconds

TIME BUILDING(B) UFO(U) RATIO (approx) image width width U/B (mm on screen) (mm on screen) 6 sec 50 30 0.60 7 50 30 0.60 8 50 30 0.60 9 48 30 0.63 9.5 45 30 0.67 10 43 30 0.70 10.5 42 30 0.71 11 40 30 0.75 11.5 38 30 0.79 12 38 30 0.79 13 38 30 0.79 14 39 30 0.77 15 43 30 0.70 16 47 30 0.64 16.5 50 32 0.64 17 53 32 0.60 17.5 55 35 0.64 17.6 56 35 0.63 (object disappears) If the UFO were at a constant, large distance from the camera (hundreds of meters, for example), as is the building, changes in the image size of both the building and UFO would be caused only by changes in the "zoom" factor (magnification). Therefore, if the distances were constant the ratio above would be constant. The variation in the ratio could be evidence that the UFO moved toward the camera during the "unzoom" portion of the video (8 - 12 seconds) and then away from the camera during the "zoom" portion (13 - 17 seconds). The UFO did move slowly to the left, so it is possible that it moved toward and away from the camera during the unzoom and zoom portions of the video. However, this would mean that the motion of the UFO was correlated with the unzooming and zooming of the camera, as if the UFO "knew" when to approach and when to recede.


Now consider the last image before the object disappears in a white flash.
Note that there are 7 lights shown, apparently around the edges of a circular object as seen obliquely from below and to one side. The 2 brightest lights "in front" (closest to the camera) appear to divide the left-right distance into three equal length segments. Also, they lie in a perfectly straight line drawn through the left and right lights (see below).
There are 3 lights in the back and they make a slight downward arc (not in a straight line). The immediate impression from looking at this image is that the object was circular with lights equally spaced around it. But there are two geometry problems. Can you see what they are? (Hint, or more than a hint: the arc of lights at the back is convex downward, as they should be for lights around the back edge of a circular object seen from below. The lights "in front" combined with the left and right end lights should also make an arc, but they should make an arc that is convex upward. They shouldn't lie in a straight line.) Now consider the spacings of the lights. The image below shows the horizontal pixel locations of the light images.
From 103 to 142 is 39 pixels and from 180 to 221 is 41 pixels. Assuming this is a round object with 7 equally spaced lights, the 39 pixels at the left side (front lights) and 41 at the right (front lights) imply that there has been a slight rotation clockwise as seen from above (each front light shifted 1 pixel to the left) from the rotation angle that would give a perfectly symmetric view. From 103 to 159 (back center) is 56 pixels and from 159 to 221 is 62 pixels. The 56 pixels at the left and 62 at the right of the "center" back light imply a slight counterclockwise rotation (3 pixels to the left for the rear light). Clearly there cannot be both a clockwise and counterclockwise rotation. This suggests that the light images are not evenly spaced around a circle. This can be further checked by using a circle as a model for the bottom of the UFO and determining where the light images would be if seen in a perspective view such as in this video. The table below shows the pixel locations of the lights in the picture and, for comparison, the locations IF the lights were equally spaced on a circle and if the viewing angle provided a perfectly symmetric view of the lights around the UFO, with two of the lights nearly at the left and right ends of the circle (as seen nearly edgewise or as a thin ellipse). LIGHT LOCATION ACTUAL PIXEL SYMMETRIC (reading left to right) LOCATION LOCATION First light (left end) 103 103 Second light (rear) 125 116 Third light (front) 142 135 Fourth (rear) 159 162 Fifth (front) 180 189 Sixth (rear) 199 208 Seventh (right end) 221 221 Obviously the locations of the lights that are between the end lights do not match up well with the locations expected if this were a symmetric circular object. A possible slight rotation (1- 3 pixels) of the UFO either clockwise or counterclockwise will not account for this magnitude of disagreement. CONCLUSION: This video is suspect because the UFO image focus does not change as much as the building focus (the focal changes should be the same if the UFO were far away, like the building). It is also suspect because the UFO image size does not change as much as the building image where the camera "zooms" in or out. And finally, the "construction" of the UFO image does not seem consistent with what might be expected from a symmetric circular craft. Of course, UFOnauts can make unsymmetric vehicles, one supposes, so perhaps this is not as important as the issues related to defocus and magnification change. Without further information that could provide reasonable explanations for the "fingerprints of a hoax" this video has to be considered a probable hoax.