Re: Camera software

--- In CAD_CAM_EDM_DRO@yahoogroups.com, "Dennis Schmitz"
<denschmitz@...> wrote:

>
> I thought of that, but I don't know enough about laser speckle to
>make it work. I understood that much of laser speckle is in the
>optics of your eye and is illusory in that it isn't really
>originating where it looks like it's originating. As I understand it,
>the light is really coming from the center of the spot and the
>interference pattern from the surface texture there
> causes the speckle to spread out once it hits your retina.

Speckle is caused by interference, if you take two plane waves and
interfere them you get fringes, if you take a large set of plane waves
that are going in all different directions you get the superposition
of lots of fringe patterns, that is the speckle.

The size of the speckle you (or any other imaging system) sees is
dictated by the range of angles that the plane waves can have at the
detection plane (retina or CCD array)and this comes down to the
optics. Specifically it comes down to the lowest NA (numerical
aperture, a measure of the range of angles) seen in the system, this
may be due to the roughness of the glass or it may be down to the
optics themselves.

To confuse matters further you can produce speckle by shining light
through a diffuser and allowing it to project onto a screen or you can
look at the back reflection from a rough surface. In the former case
speckle size is a function of the roughness and distance between the
diffuser and screen and in the later it is a function of the NA of
your eyes as well as the range of angles in the reflected light.

>But then I ran up against the biggest problem -- that a bright enough
>laser illuminator to be seen in daylight would not be eye safe, and
>if you put a tight-band filter on the imager, you can't use it for
>anything else.

It would come down to the object size I guess, a 5mW laser pointer
could be easily seen with only slightly subdued lighting. The trick
of taking an image with and without illumination before doing a
subtraction might also help.

>Besides, the idea is to have a passive system using cheap
>off-the-shelf parts -- the white noise projector is to get out of
>trouble in case we have a lot of problems with flat, featureless
>surfaces.

You are quite right, easy to get carried away. One further thought
might be to use a strong white LED or halogen bulb and project a
shadow pattern on the object. This might require surface preparation
unfortunately which rather takes away from the simplicity of the system

Graham