48 
calibrate before each new project, 25 per- 
cent once a year, 12 percent once after 
purchase of equipment, and the remain- 
ing ones have no fixed pattern. 
(6) Stability of Calibration: As this is an indi- 
vidual characteristic for each camera, the 
answers vary from insignificant 
changes” to changes of up to 0.1 mm be- 
tween calibrations. 
(7) Stereo Coverage: Only 13 percent do not 
require stereo coverage, while 32 percent 
use two cameras, 26 percent move the 
camera, 23 percent move the object, and 6 
percent use mirrors. 
(8) Control Requirements and Accuracy: 
This question was rarely answered, with 
the figures ranging from 4 to 20 points per 
model, and the accuracy (depending on 
the project) from 0.01 mmto 1 cm inobject 
space. 
(9) Evaluation: Here a 50-50 split between 
graphical and numerical evaluation is re- 
ported. 
(10) Compilation Instrument: 50 percent use 
ananalogue plotter, some solely for num- 
erical purposes; 40 percent a comparator; 
and 10 percent other devices such as rec- 
tifiers, projectors, or analytical plotters. 
Finally, it should be noted that in each reply 
oneormore specific applications were quoted 
which cover a variety of fields, such as plant 
growth and human forms, recording and 
movement studies in architecture and art, 
mining tunnels, quarries, moving objects and 
model experiments, structural and machine 
part deformation, forest inventory, water re- 
source and ocean wave studies, accident in- 
vestigation, vehicle speed control, wearing of 
road surfaces, and electron microscopy in 
chemistry and metallurgy. 
These data, my own research experiences, 
additional literature, and numerous discus- 
sions with colleagues at national and interna- 
tional meetings, especially the ISP Commis- 
sion V Symposia in the U.S.A. (Washington, 
D.C. and Urbana, Ill.) provided the informa- 
tion, with which I now shall try to respond to 
the program ofthe working group as outlined 
before. Obviously a certain personal bias may 
be apparent, but this might serve to initiate a 
discussion. As some ofthe topics will be dis- 
cussed in detail during the Congress through 
invited papers and a panel discussion, I shall 
not interfere by reporting on them here. Fol- 
lowingthe individual program points, I would 
like to make these comments: 
To No. 1) 
A non-metric camera is a camera whose 
interior orientation is completely or partially 
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1976 
unknown and frequently unstable. All “off- 
the-shelf" or "amateur" cameras belong to 
this category and are perhaps rather easily 
classified by the lack of fiducial marks. Con- 
trary to common opinions, the term non- 
metric does not imply any quality statement 
and has as such nothing to do with accuracy, 
information content, or other such charac- 
teristics. Interior orientation in this context 
encompasses the basic parameters principal 
point and principal distance or calibrated 
focal length (camera constant) as well as radial 
(symmetric) lens distortion, decentering (fre- 
quently considered in form of its components 
asymmetric and tangential) lens distortion, 
film deformation, and affinity. 
Classification of non-metric cameras ap- 
pears to be impractical, as shortcomings can 
be counteracted by more sophisticated com- 
puter usage. Perhaps the stability of interior 
orientation might act as a means of classifica- 
tion. 
To No. 2) 
The invited paper by Dr. O. Kolbl, Switzer- 
land, entitled “Accuracy Aspects Concerning 
the Interior Orientation of Non-Metric Cam- 
eras’, which follows this report will form a 
solid base for further discussions on this prob- 
lem area. 
To No. 3) 
The questions concerning object space con- 
trol are very much project and equipment 
oriented. Itis, therefore, extremely difficult to 
provide a general answer. Since users of met- 
ric cameras are also faced with similar prob- 
lems, an answer may arise from the various 
discussions in all Commission V sessions. 
Items 4 to 6 will be covered in detail during 
the 2nd session of the working group at the 
Congress, where the invited paper by Mr. 
VanWijk and Dr. Ziemann (Canada) entitled 
“The Use of Non-Metric Cameras in Monitor- 
ing High Speed Processes” will be followed 
by a panel discussion on the “Actual Use of 
Non-Metric Cameras in Photogrammetric 
Practice”. Of the seven experts, two are as- 
sociated with photogrammetric instrument 
manufacturing companies in order to insure 
the representation of metric cameras and pro- 
vide a basis for comparisions between them 
and non-metric ones. 
Concluding, I would like to state that the 
use of non-metric cameras has expanded 
within the past four years and has made an 
impact in a large number of areas where mea- 
surements are required. The non-metric 
camera/computer evaluation combination has 
reached its fullest potential, and accuracies 
reaching the photogrammetric noise level 
have been achieved. It often depends on the