A TIGHAR member who is a photographer emailed us to say that he was "distressed by this post by Jeff Glickman." He felt that, "There are so many things wrong here that makes quibbling over one-inch to be rather laughable." He went on to raise several questions. "The photograph of Amelia Earhart was apparently taken with a twin-lens reflex camera held at waist height. What was the height of the photographer? Was the camera at his belt buckle or belly-button height? How far from Amelia was he? ... The Canon 5DSR is a "full-frame" digital camera, so to get the same perspective as an 80mm 120 camera, the zoom lens should have been set to 50mm, or slightly less. Was it?" He was also concerned that the Vega in the Smithsonian is on a jack-stand. "What compensation was used to calculate the offset from the airplane's angle shown in the original with the angle in the museum photo?"
"In short, there are too many variables not accounted for between the two photos to place any reliance on the difference of one-inch in height of Amelia Earhart."
In conclusion he wrote, "I honestly look forward to a refutation of my few amateur details, but I honestly question this report's conclusion as to her height. That darned trigonometry keeps getting in the way."
His email was an honest, civil, and serious critique by a TIGHAR member with expertise in photography. We forwarded it to Jeff Glickman with a request that he respond. Jeff agreed and his reply is shown below. I asked Jeff for permission to post his reply here on the Forum (with the name of the addressee redacted) in the hope that it will remind everyone that TIGHAR is fortunate to have the pro bono help of experts of Jeff's caliber.
Dear R. ,
Thank you for your concern about the methodology that I summarized in my report regarding the Forensic Determination of Amelia Earhart’s height. I am a fellow photographer of 50 years. You are of course correct in stating that this is mostly about trigonometry. However, the handling and processing of this trigonometry is complex. Because of this, there is a subfield in photography that is dedicated to this subject which is called “Photogrammetry”.
I would first like to share with you some of my background. As a photographer, I have used nearly every class and type of camera. I have developed my own film, both black and white and color, and printed my own photographs. I have built many optical systems, including camera systems from scratch, and I hold patents in optical systems. Over more than 30 years I have developed a specialization in forensic photogrammetry and photointerpretation, as I have a deep understanding of optical systems, including cameras, and how they form and record images. In this capacity I have provided forensic photogrammetry services to the commercial sector, scientific organizations, and local and federal law enforcement, including the US Attorney’s Office, the US Department of Justice, the US Department of Homeland Security, and others. From a credentials perspective I am a Senior Member of the Institute of Electrical and Electronics Engineers, Past President of the Seattle Chapter of the Institute of Electrical and Electronics Engineers, President of the Puget Sound Region of the American Society of Photogrammetry and Remote Sensing, a Board Certified Forensic Examiner, a Fellow of the American College of Forensic Examiners, and a Past Member of the State of Washington Forensic Investigations Council.
Photogrammetry has been around more than 100 years, and there are thousands of practicing professionals in the United States represented by the American Society for Photogrammetry and Remote Sensing. There is an incomplete introduction to photogrammetry available on Wikipedia[1]. For a more thorough introduction to photogrammetry I recommend “Introduction to Modern Photogrammetry” by Edward Mikhail[2]. Also valuable to read is “Digital Photogrammetry” by Toni Schenk[3]. For a complete description of photogrammetry please see the “Manual of Photogrammetry 6th Edition”, a 1,318 page tome[4]. Recently I have been invited by a scientific publisher to write a definitive text on forensic photogrammetry.
I would like to address your concerns one by one.
What was the height of the photographer? This variable affects the height of the camera which is discussed immediately below.
What was the height of the camera? The height of the camera can be important. In this case however, the horizon is visible, and Earhart is standing erect, which tells us that the photographer held the camera substantially perpendicular to Earhart. The photograph was measured for both barrel and pincushion distortion, both of which were absent. This means that there is virtually no distortion to the length of Earhart observed in the photograph.
How far from Earhart was he? The methodology that I employed is independent of this distance variable. Therefore the distance of the photographer from Earhart does not need to be known.
Does the camera type matter? The type of camera used for the historical imagery and the reference imagery does not matter. The optical principles apply equally to both. Both images must be tested and corrected for any distortions, including lens distortion, and rotational errors.
Does the focal length of the lens matter? Yes. A long lens will distort less than a shorter lens. These distortions were detected and corrected in the paper.
Does the angle from the rear skid through the main wheels matter? This depends on the method used to develop the scale relationship in the photograph. Raising the rear skid causes the engine, propeller and cowling to tilt downward. If a vertical scale is used in or around these components, the tilt will causes an error in the calculation. The method used in the paper uses a horizontal scale to circumvent this problem, because the vertical and horizontal distortions introduced by a pure vertical tilt are independent dimensions, resulting solely in anamorphic vertical compression. This method only works, because a substantial amount of effort was made to recreate the oblique angle with which the historical photograph was taken, within the reference photograph.
Do the shoes worn by Earhart matter? Yes. As I said in my report, the heels cannot be directly measured in this photograph. Most women’s shoes from this period appear to have had 1” or 2” heels[5]. However, the imagery does not support this as there are no visible shoelaces in the left-hand shoe, and the toes are curved upward. Earhart was known to wear boots and the limited evidence in the photograph suggests that she was wearing boots in the photograph. Boots sometimes have shorter heels which would explain the upward curved toes. In my opinion the shoes may have ½” heels. You may substitute a different amount if you wish.
Does the hair compression seen on Earhart matter? Yes. As I said in my report, there is little in the scientific literature that describes the height of hair above the top of the head. My best estimate based on circumstantial information is ½”. You may substitute a different amount if you wish.
Needless to say, photogrammetry is a complex topic. If it were simple, we wouldn’t need a 1,318 page to explain it. I would be happy to instruct you in its intricacies and subtleties should you wish. If you have any additional questions, please feel free to call me. I would be happy to share with you my decades of experience working in this highly specialized area.
Sincerely,
Jeff B. Glickman, BSCS, BCFE, FACFE, DABFE
PHOTEK – First in Forensic Imaging
President, American Society for Photogrammetry and Remote Sensing, Puget Sound Region
Chair Emeritus, Institute of Electrical and Electronics Engineers, Seattle Section
Senior Member, Institute of Electrical and Electronics Engineers
Board Certified Forensic Examiner
Fellow, American College of Forensic Examiners
Diplomate, American Board of Forensic Examiners
Member Emeritus, State of Washington Forensic Investigations Council
[1]
https://en.wikipedia.org/wiki/Photogrammetry[2] Mikhail, Edward, et. al., “Introduction to Modern Photogrammetry”, John Wiley & Sons, New York, 2001.
[3] Schenk, Toni, “Digital Photogrammetry”, TerraScience, Laurelville, OH, 1999.
[4] McGlone, J. Chris, Ed., “Manual of Photogrammetry 6th Edition”, American Society of Photogrammetry and Remote Sensing, Bethesda, Maryland, 2013.
[5] Email correspondence, Suzanne Petersen, Collections Manager, BATA Shoe Museum, Toronto, Canada.
