1978 ISP COMMISSION V INTER-CONGRESS SYMPOSIUM - STOCKHOLM SWEDEN 
e -— mum mem a — — wem mem Uem qum eM umm qum a que mum Wem A AMAA em deem MA an m wee cmo o m m m m 
——— — 2 2 be nn 
Pryputniewicz: on the object which we study, and the point on the holo- 
gram through which we observe. The line going through these two points 
will be the direction of observation. So we have to know the co-ordinates 
of these two points, and also we have to know the fringe shift, how many 
fringes pass the point of interest on the object as we change our direction 
of observation from one point on the hologram to another. Those are the 
only things we have to know. 
Doyle: How do you actually measure these? 
Pryputniewicz: The hard way to do it is to observe it with your eyes. 
This is a very inaccurate way of doing it, because the analysis depends on 
the fringe shift, and the more accurately you can determine fringe shifts, 
that is whether it is 1.1 or 1.2, the more accurate your results will be. 
Now Dr. Ek of the Institute of Optical Research here in Stockholm has de- 
veloped an image scanning system which gives fringe shifts down to 1/100 
of a fringe. With our technique we use a small scanner and can get fringe 
shifts to 1/10 of a fringe, and this gives usan error of about 4 or 5$. 
Q^6t'9g^g 6 $5658 $58 0$ 99 $5959 8 $556. 
Paper by L. Ek and K. Biedermann - Inst. of Optical Research - Stockholm 
"HOLOGRAM INTERFEROMETRY OF MECHANICAL DEFORMATIONS" 
* + - * . . = . - *. - > . + - * Ld - * > - * LÀ * . . Ld e * . e - * *. hd *. * 
Discussion of the above paper 
  
Gates: Perhaps I could ask a question? Could you give us some idea of the 
time it takes to make this kind of experiment? 
Ek: The set-up we have used so far has been very very primitive, so it 
has taken a rather long time. What one should do is to feed the digitized 
fringe signal into a minicomputer, for instance, and then the least squares 
fit could be made directly. But so far we have not had access to a com- 
puter, so we taken the output signal to an xy-plotter. Then I have fed the 
numbers, by hand, into the computer. Due to this, it takes time, and it's 
no fun at all. 
Gates: It is very much to your credit that you have persevered. 
  
Doyle: I believe I understand what you are doing in the case of a static 
Situation, such as the deformation of a beam. But if you have a vibrating 
object, as you illustrated first, then I do not see why the fringes are not 
moving continuously. How do you localize these moving fringes for measure- 
ment or counting? 
Ek: If you have a vibrating object then you get Bessel Function Fringes, 
which are stationary in time. When the direction of the vibration is not 
directed toward you, that is if the vibration amplitude contains at least 
a component which is perpendicular to the direction of observation, then 
the fringes will move in the same way as when you have a static displace- 
ment. The only principle difference between a vibration and a static dis- 
placement is that in the case of a static displacement you have cosinu- 
soidal fringes, and in the case of a vibration you have Bessel Function 
Fringes. But the evaluation principle is very similar. 
Gates: I think perhaps it may not have been understood, and I am not sure 
whether you said it, but of course with the vibration you are averaging over