Quantitative Evaluation of Photo 
Interpretation Mapping* 
HAROLD E. YOUNG 1 and 
ERNEST G. STOECKELER 2 
Abstract: The accuracy of maps prepared by photo interpretation 
techniques has often been questioned by those who make the maps as well 
as by many users. In the past five years a number of American workers 
in the field of photo interpretation have emphasized the need for quantita 
tive evaluation of photo interpretation in Photogrammetric Engi 
neering. 
To illustrate the type of information required and the analytical 
procedures necessary to make a quantitative evaluation, an engineering 
soils map of a 210 square miles area (544 square kilometers) in the 
State of Maine was used as an illustration. Data were presented com 
paring photo interpretation classification versus actual field classifica 
tion of soils. By applying recognized statistical procedures, the authors 
prepared a table presenting photo interpretation errors based on field 
sampling at 19:1 odds. These data were further refined to show the error 
at 19:1 odds within soil types. In addition, a hypothetical illustration 
was presented to show the method of determining the accuracy of bound 
ary line location between adjacent soil types. 
In the authors' opinion, most mapping studies of the earth's surface 
and vegetation should be evaluated quantitatively so that the users can 
employ the map information more intelligently. With an accumulation 
of results of quantitative evaluation of a number of photo interpretation 
studies, it will be possible to compare photo interpretation accuracy 
resulting from such controllable factors as scale, season, film and filter 
as well as differences between photo interpreters. 
Introduction 
F orest type maps, engineering soils maps, maps of geologic studies, geo 
graphic studies and terrain analysis maps for military purposes of hundreds 
of thousands of square miles of territory have been prepared with the aid of 
photo interpretation techniques. It is inevitable that many people employing 
the maps will ask the question—how accurate is the photo interpretation? 
In most instances, it is not feasible to field check more than a small portion 
of the total area, and in some cases, it is not possible to do any field checking at 
all. Thus the concern over map reliability is very real. In the past five years 
Colwell (1), Lundahl (4), Sammi (8), Katz (2), Rogers (7), Pomerening and 
Cline (5) and Young (10) have emphatically expressed the need for quantitative 
evaluation of photogrammetry and photo interpretation. 
The papers and comments cited above should have aroused some interest 
among those who are doing photo interpretation of the earth’s surface and 
vegetation. Krumbein (3) and Potter and Siever (6) have recently employed 
* A Contribution of Commission VII, Photographic Interpretation, International Society of 
Photogrammetry. 
1 Assistant Forester, Maine Agricultural Experiment Station. 
2 Research Assistant, Soils Mechanics Laboratory, University of Maine.