Full text: Precision and speed in close range photogrammetry (Part 1)

UA 
INTRODUCTION 
In the ideal model for close range photogrammetry, each camera is 
assumed to function like a distortionless pinhole camera, which projects 
all the required bright target-points inside the volume under study into 
sharp recorded images on a flat photographic plate. In designing lenses 
for this purpose great care is taken to minimise distortion, but it is 
difficult to control the inner and outer centres of perspective, and a 
conventional lens does not give good depth of focus simultaneously with 
sharp lateral resolution. 
A new camera is being developed at NPL which overcomes both of these 
difficulties. Firstly, freedom from distortion is ensured by adopting 
a monocentric lens design, for which all the refracting surfaces are 
spherical and concentric. If we consider any principal ray, which passes 
from an object-point through the common lens centre and then on to the 
image-point, it is readily seen that the path of this principal ray must 
be a straight line, since it encounters every surface at normal incidence. 
See Fig. 1, in which the path of each principal ray is shown dotted for 
three different object directions. 
The second problem to be faced is that of ensuring that all of the 
images recorded by the camera are not only symmetrical about the principal 
ray but also sharp enough to be measured accurately. This sharpness must 
be maintained for all object distances from infinity down to perhaps 500 mm, 
and for off-axis regions as well as for central regions of a nominally 
flat photographic plate. As shown in Fig. 1, this implies rod-like images 
which remain sharp over at least 15 mm depth of focus (assuming a 71 mm 
plate distance). Fortunately enough, images of this kind can be generated 
by designing a monocentric lens with strong negative spherical aberration. 
AXICON IMAGING 
As illustrated in Fig. 1, negative spherical aberration has the effect 
that, as seen from one of the image-points on the recording layer, the 
converging wavefront is no longer spherical, being excessively concave in 
its central region but increasingly convex toward its periphery. There is 
however, a narrow annular region for which the wavefront normal, and hence 
the associated ray-direction, points directly towards the image-point. It 
ai 
 
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.