Considerable progress is being made in the improvement of more 
or less conventional techniques through innovation and evolution 
that may, in fact, surpass in usefulness and interpretive 
benefits, the possible benefits or fall outs from present 
efforts toward development of the fully automatic photo 
interpretation system. 
From these general statements, one of two conclusions may be 
drawn. Either 1) the problem of a fully automated system is impossible to 
solve, or 2) the problem is not fully understood. 
It occurs to this reporter that the answer lies somewhere between 
the two. Some aspects of the problem are impossible to solve and to demonstrate 
this let me again quote the definition of photo interpretation as given by 
(14) 
the Photo Interpretation Manual published by the American Society. 
"Photographic Interpretation - The act of examining photographic images for 
the purpose of identifying objects and judging their significance." As 
all workers in this field will admit, no one is attempting to automate the 
second part of the definition that deals with the judging of the significance 
of objects. The problem arises, however, that in attempting to automate 
the identification of objects, it is often overlooked that identification, 
particularly in an aerial photograph, is dependent upon judging the signi- 
ficance of a number of factors that have influenced the image which is trying 
to be identified. 
The result is that primitive, high contrast, laboratory prepared 
targets can be automatically interpreted as can the interpretation of selected 
(14) "Manual of Photographic Interpretation", American Society of Photo- 
grammetry, 6269 Leesburg Pike, Falls Church, Va., 1960.