AN ANALYSIS OF LENS PERFORMANCE 
by 
Duncan E. Macdonald.- 
The particular work discussed is the product of the aerial photographic 
research program at the Boston University Optical Research Laboratory. This 
work has been sponsered by the U.S. Air Force Photographic Reconnaissance 
Laboratory. 
The Boston University program is based on the concept of an aerial 
photographic system shown in Figure 1. This concept is based on the fact that 
Fig. 1. The Aerial Photographic System. 
the purpose of aerial photography is the transmission of a message which 
passes through the various components of the system. Each box in this photo 
graphic system tends to degrade the information content of the message by 
introducing some distortion or some noise. 
In the following, we shall concern ourselves with assessments of relative 
performance of lens-film combinations. 
It is well known that any lens-film combination will reduce the image 
contrast more and more as one goes toward finer detail. The nature of this 
reduction is shown in Figure 2. The performance of three aerial photographic 
lenses employed in conjunction with Super-XX emulsion has been examined 
on axis for the case of a typical resolution target of alternating white and 
black lines of equal width, and for the case of an isolated white line on a black 
background. 
It is in order to insert that the particular illustration used here and 
throughout the paper neglects the role of exposure level and the effect of shape 
of the recorded symbol. However, the particular functions shown are held to 
be representative of performance of the aerial photographic system, first, 
because the brightness range has been selected to cover the maximum condi 
tions encountered in operations, and, second, this has been done at the most 
representative exposure level. 
There is a level of contrast below which the human observer is not able 
to detect a symbol on a given photographic material. There is much in litera 
ture to guide us in the selection of the visual contrast threshold \ In addition, 
there exists some laboratory experience for interposing the effects introduced 
by the inhomogeneities of the photographic material. From this, we assume 
a reasonable threshold value. The data of Figure 2 are then used to determine