The Panel recommends that, given a new alleged UFO photograph, the decision to invest effort into its investigation should be taken only if both of the following conditions are fulfilled:
If, after visual examination, the displayed object has not been identified (planet, balloon, cloud, etc.), investigation should be performed in two steps:
When the original film is available and analysis seems justified, all technical data concerning the site, viewing conditions, camera, film, processing, etc., must be collected. If the camera is available (in an ideal case still loaded with the original film), it must be used to perform the following calibrations:
The film should be processed under rigorously controlled conditions (if it has not already been processed commercially). If the camera is available but empty, the same operations should be conducted with a film of the same type as the original.
The investigator should visit the original site and make measurements concerning the three-dimensional geometry of the observed landscape or this information should be extracted from detailed maps. If the photograph has been acquired at nighttime, an astronomical map of the sky at the time of acquisition will be necessary. The investigator should determine the meteorological conditions from the official offices or air bases in the neighborhood with particular attention to the horizontal visibility. The investigator should also take into account all quantified or quantifiable elements of the witness testimony including the estimated shape, angular size, velocity, color, etc.
For analysis of the photograph, it is essential to work from the original negative. This should be carefully washed and examined under a microscope to look for possible tell-tale artifacts and scratches, and to check the regularity of the grain structure so as to detect multiple exposures. The negative should be analyzed by conventional photographic instruments (enlarger, projector, etc.), and the information on the negative should be digitized by a microdensitometer.
Once digitized, the image may be analyzed by computer analysis, using the classical tools of contrast enhancement, noise suppression, contour detection, restoration, etc., and more specialized techniques such as maximum-entropy analysis that may be used to remove the effects of target motion and/or camera motion. Such analysis will assist in the detection of a possible hoax. For instance, a suspension thread may be brought into evidence through standard differential operations. Also, one may estimate the distance (hence the size) of the object through MTF computations, based on an analysis of atmospheric diffusion and contour blurring. If there are black areas on the object, it is possible to obtain estimates of the distance by comparing the luminance of such regions with other identified black parts of the scenery. If the object is nearer than the minimum depth of field, one should be able to detect geometrical distortions in the image. If the operator had a slight movement while taking the picture, analysis of the corresponding blur on the object and on other elements of the landscape may allow the calculation of a possible range for the distance of the object.
In the case of a color photograph, one should carry out the above procedures in three steps using three appropriate color filters for scanning.
If an event is recorded on a cine camera, each frame may be analyzed as above. However, it is now possible to obtain additional information by combining and comparing the sequence of images.
In principle, images recorded by video cameras may be subjected to comparable analyses. However, video records suffer from one very important weakness: since the basic data is in electronic form, it could have been modified by the use of suitable electronic equipment, so that the authenticity of a video record will depend even more critically upon the credibility of the witness testimony.
