Computer Analysis - July 1990 - David Cook

Analysis #: MU-001 
Reference: August 23, 1987  10:30 PM Corydon, Indiana
Contact:    Francis L. Ridge, 618 Davis Drive, Mt. Vernon, IN  47620 
Media:      VHS Video Tape, generation unknown 
Result:       Inconclusive

          The images, as they appear on the video tape, are too small to provide conclusive evidence as to their shape and coloration.  The background imagery was darker than the black-level capability of the camera, therefore no measurement of the size or relationship to background objects was possible. Camera movement added extreme motion blur which further inhibited analysis.

          The videotape allegedly shows three objects. The first is supposed to be a plane, near an airport.  The second and third objects are unidentified. The analysis focused on the first and second objects, as the third was too distorted and small to provide adequate sampling. The first object was analyzed as a control to the second object.

Object #1:
          The first objiect on the video tape was sampled at 756 x 482 with 32 bits per pixel, from moving video tape. The signal was converted from NTSC to RGB prior to sampling. Seven samples were taken, at various intervals, and one of the seven samples was chosen for analysis due to its clarity. Techniques incluide:

Motion Analysis:
          This sample has extreme motion blur caused by camera movement. in the.1/30th second frame, almost no non-camera movement can be detected due to the extreme camera movement.

Shape Analysis:
          This image takes approximately 10 scanlines by 15 pixels and consists of four very bright lights arranged in a delta formation. Noise structures, which are brighter than the background noise, attach the two outermost lights near the front light. Conclusion is that this is some type of aircraft.

Color Analysis:
          Due to the oversaturation of the image against the blackness of the background, coupled with the motion blur, colors are not conclusive. However, the front light is much brighter and more pure than the other lights, being mostly white. The 'wing' and 'tail' lights are all reddish. There appears to be a green cast under the front 'white' light. Conclusion is that the front light is white but the coloration of the other lights is distorted due to oversaturation and lens dynamics.

Object #2:
          The second object on the video tape was sampled at 756 x 482 with 32 bits per pixel, from moving video tape. The signal was converted from NTSC to RGB prior to sampling. Over 10 samples were taken, at various intervals, and two of the 10 samples were chosen for analysis due to the reduced motion blur on these samples, Techniques include:

Motion Analysis:
          Both images contain motion blur in two directions, One of the motion blur streaks is large and uneven, the second streak is small and even. The large streak is to the horizontal left of the objects and the smaller steak is to the upper right of the first sample and to the lower riqht in the second sample. Conclusion is that the large streak is motion blur due to camera movement and oversaturation of the vidicon from frame to frame (this is confirmed by the fact that the left blur exists on multiple consecutive frames while the riqht blur changes on consecutive frames). The conclusion drawn from the right streaks is that they represent motion of the object, in a period of l/30th of a second. This rules out stars.

Shape Analysis:
          Both samples show the object only taking five (5) scan lines on the raster. The relative X size is approximately 10 pixels, which qives a total area coverage of 50 pixels and represents 0.00013 of the total viewable screen. This amount of area is too small for detailed object analysis, however, the following observations were made. Both samples show a oval shape with a wider 'band' in the middle. The top portion of the oval (above the band) is larger than the bottom portion of the oval (below the band). In one of the samples, the band is quite visible and extends into the left side motion blur. In both images, the bottom portion of the oval appears much more tapered than the top portion.

Color Analysis:
          Because of the extreme smallness of the images and their brightness in contrast to the absolute darkness of the background, color analysis is also inconclusive. However, the following observations were made. The second sample shows fairly even coloration with slight green and red tinting. The first sample shows mutch more coloration with a green base, a reddish-purple to orange midsection and a bluish top. Due to the difference in both samples, the conclusion is that any coloration is due to oversaturation coupled with motion blur and lens dynamics. 

Background Analysis:
          The background of both samples were enhanced without the object to see if any backqround detail was recorded by the film which could be used to establish speed and distance. Enhancement of the images produced only banded noise and no objects can be seen. This is due to the fact that the black-level of the camera is higher than the blackness of the night, in contrast to the bright object. Only noise was recorded.