Searching \ for '[EE]: Multiple small object detection' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=multiple+small+object
Search entire site for: 'Multiple small object detection'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Multiple small object detection'
2001\01\22@013740 by Anthony Bussan

flavicon
face
I am trying to build a system that finds the X-Y position of about 400
pieces of lead shot (BB's, balls) as they pass through a 4 foot by 4 foot
"window."  The shot is approximately .040 inch diameter and is flying at
1200 feet per second.  They all come through the window at pretty much the
same instant (this is from a shotgun).  I really don't know where to start
on this.  I did some searching on the web, but I didn't find anything close
to this level of complexity.  Do any of you guys(or gals) have any insight
on where I should start researching this?  I don't know what kind of sensors
or electronics could handle this kind of speed, size of object, and volume
of data all at the same time.
Thanks in advance!
Tony

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@040217 by spam

flavicon
face
Hi Anthony,

You are saying X-Y and not X-Y-Z, so time is of less consequence.
You could try setting up 3 line scan sensors with optics (in a
triangle) with a bright light behind each, shaded so that you don't
get light directly onto any of the other sensors. You will then get a
fairly good reflection from each ball.

If you open the exposure gate of the CCD when the shot goes and
expose for a few ms, you will be sure to get a picture of the ball
paths .

You might be able to time a pulsed array of infrared LED's. Just put
3-4 in a row without resistors in front and give them a good 10A
pulse with a fet (that would be 12V). Plenty of light for about
500uS, just enough to catch the whole shower if it moves at about
300 m/s. (??)

Now you have 3 projections of the showers. To avoid too much
math, just divide the window in a grid (could be 128x128) and
check - IF there was a ball passing here, we would have light in
LCD1[x],LCD2[y],LCD3[z].

If there is light all three places, assume a ball passed here.

You might get a few extras, depending on the resolution.

If you use 4 sensors, you can use a polarizer in front of light + ccd.
Cross them, and the CCD won't be able to see the light shining
right into it. It costs you half of the light, but hey, good infrared
LED's cost about 50 cents.


Hope this gives you an idea where to start.

Kent

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@090524 by Olin Lathrop

face picon face
> I am trying to build a system that finds the X-Y position of about 400
> pieces of lead shot (BB's, balls) as they pass through a 4 foot by 4 foot
> "window."  The shot is approximately .040 inch diameter and is flying at
> 1200 feet per second.  They all come through the window at pretty much the
> same instant (this is from a shotgun).  I really don't know where to start
> on this.  I did some searching on the web, but I didn't find anything
close
> to this level of complexity.  Do any of you guys(or gals) have any insight
> on where I should start researching this?  I don't know what kind of
sensors
> or electronics could handle this kind of speed, size of object, and volume
> of data all at the same time.

Does the device have to work more than once?


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, spam_OUTolinTakeThisOuTspamembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@092602 by Simon Nield

flavicon
face
olin:
>Does the device have to work more than once?

you're thinking "big sheet of cardboard" ? think i have seen a similar technique used before ... :)

Simon

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@100719 by Bob Ammerman

picon face
How about a big sheet of paper/plastic that is then post-processed by a
scanner?

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@103439 by Stephen B Webb

flavicon
face
> I am trying to build a system that finds the X-Y position of about 400

So you will be "lookin' down the barrel of a gun"?  Or looking from the
side as the shot travels parallel to the X-Y plane.

Do you *really* need to have the X-Y position of each piece, or would some
sort of approximate density cloud image work?

Assuming you are looking down the barrel, you could get a high speed
digital camera to image it, I would think.  Looking from the side requires
quite a bit higher shutter speed to "freeze" the shot in mid flight.

I'm just afraid of the resolution requirements, though.  Even if 1 pixel
in the image were enough to represent one piece of shot, you would need a
minimum of  (4*12)/0.040 pixels in both the horizontal and vetical
(1200x1200), but the more pixels the better.  So, I would see if you could
reduce your 4' x 4' area a bit.

www.spectraservices.com/digital/cooke_hsfc_pro.html
www.rit.edu/~andpph/text-agfa-1280-hs.html
http://www.fillfactory.com/htm/track_record/htm/high_speed.htm

How accurate do the measurements need to be?  The easy way would be to
point the camera at the opening but not in the path of the shot, and
calibrate the camera to the opening with a homography.  If you really need
it "right", then you will have to calibrate the lens to recover the radial
distortion, etc.  With a high end lens, you may find the radial distortion
to be low enough that it can be ignored.

Once you acquire the image, a few image processing tricks may be able to
pull out the coords of all of the pieces of shot.  (Would it be acceptable
to only recover 300 of them?)  Would it be acceptable to have to sit in
front of a computer and hand pick them?  Is this whole setup already too
complicated?

> or electronics could handle this kind of speed, size of object, and volume
> of data all at the same time.

The above links address some of these issues.

I believe the last one is a camera that can do 1k x 1k pixels and 1000
fps, so your shot would move 1.4 feet between frames.  The shutter speed
is high enough to avoid motion blur.  1/100,000 sec shutter speed.


I am assuming that the image processing / position recovery can occur
off-line.


-Steve

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@134659 by Dale Botkin

flavicon
face
On Mon, 22 Jan 2001, Simon Nield wrote:

> olin:
> >Does the device have to work more than once?
>
> you're thinking "big sheet of cardboard" ? think i have seen a similar technique used before ... :)
>

I know I was thinking "big array of addressable somethings" and see which
ones still exist afterward...  8-)

Dale
---
The most exciting phrase to hear in science, the one that heralds new
discoveries, is not "Eureka!" (I found it!) but "That's funny ..."
               -- Isaac Asimov

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@160057 by Philip Martin

flavicon
face
Tony,

I cant help you with the electronics for this, but you are making the wrong
assumption about the shot pattern. Its amazing how many people think it
comes out of the muzzle as a flat plate of shot or as a ball. In fact the
shot pattern is a long cone type pattern, often up to 4 to 6 feet long.
That's how come we can bust a clay on the move, so long as the clay goes
through the pattern at some point, usually!

Philip Martin.


{Original Message removed}

2001\01\22@180629 by Anthony Bussan

flavicon
face
I know you guys are probably thinking that I could just shoot at a piece of
paper and then look at it.  I do appreciate the KISS solution, but this is
already where I am at.  When you are "patterning" the shot from a shotgun,
you try to get the best distribution to optimize your chances of hitting the
clay bird.
This weekend, after patterning 6 kinds of shells, through two different
guns, with 3 different choke tubes each, and at several distances from the
target paper, I ended up with nearly 100 pieces of paper hanging in the
shop.  4 by 4 foot paper takes up a little room anyway.  Then you have to
manually count over 300 holes on each piece, for now we ignore the holes
outside of a certain diameter.  Then you look for gaps in the pattern and
subjectively judge the different combinations.
I would like to be able to electronically capture the shot pattern to speed
things up a lot, and then put together some software to analyze it and graph
it out.  Some statistical methods could be employed too.

Thanks to everyone, I really appreciate the help so far!

Tony

olin:
>Does the device have to work more than once?

simon:
you're thinking "big sheet of cardboard" ? think i have seen a similar
technique used before ... :)

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@181247 by Anthony Bussan

flavicon
face
Wow, I didn't know the shot stretched out so far.  I always thought the wad
kept the shot pretty much bunched together as it left the barrel.  Seeing
the "cone" of shot would be interesting, but I think I might be digging a
bigger hole for myself if I try to capture that too.

Tony



{Original Message removed}

2001\01\22@183406 by Bob Ammerman

picon face
So, are you or are you not interested in the time domain?

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

{Original Message removed}

2001\01\22@184828 by Anthony Bussan

flavicon
face
I do not want to know when they pass through my device(I keep envisioning
shooting at/through a picture frame).  I just want to know where they passed
through.  If the Z axis is the line of travel I just want to know the X-Y
positions at the plane of the "picture frame".
As far as accuracy goes, I am thinking in multiples of the diameter of the
pieces of shot.  Maybe +-1 or 2 diameters.

Tony



{Original Message removed}

2001\01\22@203137 by Andrew Warren

flavicon
face
Anthony Bussan <.....PICLISTKILLspamspam@spam@MITVMA.MIT.EDU> wrote:

> I know you guys are probably thinking that I could just shoot at a
> piece of paper and then look at it.  I do appreciate the KISS
> solution, but this is already where I am at.  When you are
> "patterning" the shot from a shotgun, you try to get the best
> distribution to optimize your chances of hitting the clay bird.
>
> This weekend, after patterning 6 kinds of shells, through two
> different guns, with 3 different choke tubes each, and at several
> distances from the target paper, I ended up with nearly 100 pieces
> of paper hanging in the shop.  4 by 4 foot paper takes up a little
> room anyway.  Then you have to manually count over 300 holes on
> each piece, for now we ignore the holes outside of a certain
> diameter. Then you look for gaps in the pattern and subjectively
> judge the different combinations. I would like to be able to
> electronically capture the shot pattern to speed things up a lot, and
> then put together some software to analyze it and graph it out.  Some
> statistical methods could be employed too.

Tony:

I haven't been following this thread, so I don't know whether the
following has already been suggested.

   Shoot at 100 4 x 4-foot paper targets.

   Make JPEG images of each target (48" x 48" is too large to fit in
   most scanners, so take photos with a digital camera held at a
   constant distance/orientation to the targets).  Backlight the
   targets and/or crank up the contrast, so your images end up
   completely black with white dots representing the holes.

   In Photoshop, PhotoDeluxe, Paintshop, or any other decent
   image-editing program, create an image of concentric circles of
   varying colors.  Use the program's "frequency distribution
   histogram" function to make sure that the colors are equally
   represented.

   One at a time, overlay each of your target images on the
   concentric-circle image, allowing the colors to "shine through"
   the holes.

   Use the "frequency-distribution histogram" function again.

   Voila... The histogram shows the radial distribution of the shot.

   If you want to see angular distribution (i.e., you want to see
   if there's as much shot at the 3 o'clock position as at the 9
   o'clock position), divide your background image into pie-shaped
   slices instead of concentric circles.  Of course, you can combine
   the two, or use a square grid pattern or whatever.

The nice thing about this solution is that you don't have to do any
programming or build any hardware (except for maybe an easel and a
tripod); it should take less than a few hours to do the first target,
and just a couple of minutes to do each additional one.

Just an idea...

-Andy


=== Andrew Warren --- aiwspamKILLspamcypress.com
=== Staff Systems Engineer, IPD
=== Cypress Semiconductor Corporation
===
=== Opinions expressed above do not
=== necessarily represent those of
=== Cypress Semiconductor Corporation.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\22@213031 by Russell McMahon

picon face
Is invasive** measurement acceptable?
ie is the "piece of paper" idea which some have proposed potentially
acceptable.

If so - HOW invasive (eg paper versus stop-dead-on-spot substance etc.

What is the required recycle time.



     Russell McMahon


** - all measurements are invasive but I imagine quantum mechanical effects
are not of concern here.

_____________________________

What can one man* do?
Donate food daily free !!! -  http://www.thehungersite.com/
Donate Vitamin A!  http://www.thechildsurvivalsite.com/
http://www.rawa.com  - one perspective on Afghanistan
http://www.changingourworld.com    http://www.easttimor.com   http://www.sudan.com

(* - or woman, child or internet enabled intelligent entity :-))


{Original Message removed}

2001\01\22@215859 by rottosen

flavicon
face
Speed up the paper/picture process a little by using a _roll_ of very
wide paper such as used by a printing press?

-- Rich


Andrew Warren wrote:
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\01\23@050545 by Simon Nield

flavicon
face
as someone else suggested: big roll of paper (or plasticised paper, whatever survives well enough
being shot without totally disintegrating... using black paper might make your life a whole lot
easier later on...)

feed paper into one side of the window, and take it up on another roller on the other side.

shoot paper from various distances or whatever else you want to do, scrolling on roll each time.

as a second process (or as part of the same process if you like) run the paper through a scanner of
some kind...

[one resonable solution, depending on how much time/ money you want to invest might be to build your
own custom scanner (assuming you can't get a suitable one off the shelf - i have no idea about that
possibility). back-light the paper with an even(ish) source of illumination (more even lighting will
make the filtering bit later a bit easier), use a lense arrangement to focus the image onto a linear
ccd array (you can get these at very high resolutions for not too much dough) and scroll the paper
to scan the whole thing - no need to worry about linear speed really, so long as you measure the
movement and grab a scanline at the right time.
throw the resultant image through some filtering to give you a good contrast, scan picture however
you want to get the info you are after]

regards,
Simon

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@061415 by Alan B. Pearce

face picon face
> This weekend, after patterning 6 kinds of shells, through two
> different guns, with 3 different choke tubes each, and at several
> distances from the target paper, I ended up with nearly 100 pieces
> of paper hanging in the shop.  4 by 4 foot paper takes up a little
> room anyway.  Then you have to manually count over 300 holes on
> each piece, for now we ignore the holes outside of a certain
> diameter. Then you look for gaps in the pattern and subjectively
> judge the different combinations. I would like to be able to
> electronically capture the shot pattern to speed things up a lot, and
> then put together some software to analyze it and graph it out.  Some
> statistical methods could be employed too.

You do not mention doing multiple shots with a given combination to see how
repeatable the pattern is. I guess this is something to investigate once you get
the scheme up and working suitably.

{Quote hidden}

snip

I do like this as a method. I would take it slightly further for the paper
feeding in that if it was possible to get tag end rolls of newsprint to use as
the target it should not be too hard to set up a frame to hold the paper to
autofeed through the target area, but have a suitable floodlight that can be put
behind the target immediately after the shot, and then take the image before
winding on the paper. This would save the possibility of tearing the paper as it
is wound on.

Now you are going to tell me that all the shot samples are taken outside in full
sunlight. Ahh well you can not win them all, I guess some wooden frames with
suitable paper prestretched over them like artists canvas will have to do.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@154159 by Chris Carr

flavicon
face
> > This weekend, after patterning 6 kinds of shells, through two
> > different guns, with 3 different choke tubes each, and at several
> > distances from the target paper, I ended up with nearly 100 pieces
> > of paper hanging in the shop.  4 by 4 foot paper takes up a little
> > room anyway.  Then you have to manually count over 300 holes on
> > each piece, for now we ignore the holes outside of a certain
> > diameter. Then you look for gaps in the pattern and subjectively
> > judge the different combinations. I would like to be able to
> > electronically capture the shot pattern to speed things up a lot, and
> > then put together some software to analyze it and graph it out.  Some
> > statistical methods could be employed too.
>
> You do not mention doing multiple shots with a given combination to see
how
> repeatable the pattern is. I guess this is something to investigate once
you get
{Quote hidden}

use as
> the target it should not be too hard to set up a frame to hold the paper
to
> autofeed through the target area, but have a suitable floodlight that can
be put
> behind the target immediately after the shot, and then take the image
before
> winding on the paper. This would save the possibility of tearing the paper
as it
> is wound on.
>
> Now you are going to tell me that all the shot samples are taken outside
in full
> sunlight. Ahh well you can not win them all, I guess some wooden frames
with
> suitable paper prestretched over them like artists canvas will have to do.
>
In actual use the spread of the shot is also modified by the moving barrel
as you track the target. Therefore as well as knowing where the passes
through an XY plane perpendicular to the gun barrel you also need to know
when it passes through that plane.

I would suggest that the only practical way of obtaining this information is
to record the shot passing through a light curtain with a high speed camera,
digitising the results, then comes the easy bit, mathematical modification
to simulate the effect of the moving barrel.

Chris

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@175718 by Peter L. Peres

picon face
>Does the device have to work more than once?

tsk, tsk, always over-optimizing <g>.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@175741 by Peter L. Peres

picon face
>In fact the shot pattern is a long cone type pattern, often up to 4 to 6
>feet long.

Only when far enough from the muzzle afaik. A 4ft 'long' pattern would be
covered by my numbers (4 x 4 x 4 ft almost cube).

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@175808 by Peter L. Peres

picon face
Russell McMahon <apptechspamspam_OUTCLEAR.NET.NZ> wrote:
>If so - HOW invasive (eg paper versus stop-dead-on-spot substance etc.

Are you thinking large tray with filled with wax, CCD camera above it and
some heaters under it ? <G> Or was it plasticine and a motorized vertical
roller.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@175842 by Peter L. Peres

picon face
Motion blur occurs when the target moves more than 1 pixel per frame. It
has to move half a pixel or less to be sure. At 400 m/sec and a 300 pixel
CCD (vertical, we take the lower resolution of a standard TV CCD),
assuming the frame spans 80% of the explored image, at 45 degrees angle,
you get 4 ft ~= 1.2 m -> 1.2 m / (300 * sqrt(2)/2 * 0.8) ~= 7 mm/pixel.
The shot is allowed to move half that far to avoid blur which leads to
3.5E-3 m / 400 m/sec = .00000875 sec exposure/flash (about nine
microseconds, or 1/114000 sec. So you will likely take a 1/150000 sec
flash. So far, so good.

There is only one small problem. Your buckshot is 0.040 which is abt. 1mm.
Rule of thumb is, smallest object to be resolved is 2 pixels in size (per
side). By applying the first (size) calculation backwards, the result is
that you need about 14 times 300 pixels resolution (we need to resolve 1/2
of 1mm which is 0.020", instead of 7mm). This is 300 * 14 = 4200 pixels
(per side). The exposure time is also to be reduced, 14 times (for the
same reason). Thus you'd need two CCD cameras with abt. 4200x4200 pixels
and a flash with 1/1600000 sec (abt. 620 nsec) time. Because of the light
requirements the flash power would have to be gigantic (probably in the
lower Q-chopped laser range). I don't know what you are doing but I have a
feeling that this might slightly upset your budget.

On the other hand, if you accept some smear and the fact that a computer
will use the smear to compute speeds and other details (like vectors),
then you could use two standard 35 mm still cameras, probably at 45
degrees from each other and from the shot axis, and a somewhat complex
high speed high power flash (with 3-10 usec pulse) but built with standard
parts. I don't think that using 4200^2 pixel CCDs (two of them) will be a
likely option ;-). Inexpensive 35 mm cameras will certainly resolve 4200
pixels (and even ten times as much with the proper film and lens). It will
really be helpfull if you can control the moment the gun is shot (using a
remote trigger), and if the room is relatively dark. You can even use a
PIC for this, the timing requirements are within its range in this case.

I just thought I'd check my previous postings in numbers for some reality
check. ;-)

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@175852 by Peter L. Peres

picon face
2 Camera(s) + lights. If you use flashes (you should) then you can use
still film cameras (35mm) or heavyly shuttered CCDs. The resulting
stereogram can be analyzed later or in real time by a computer. You can
use one or several PICs for timing purposes and flash control. The
technhology belongs to high speed photography and imho you should look for
books on that.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@204252 by Anthony Bussan

flavicon
face
First of all, I have to say thanks to all the response I have gotten!
From what I have read, it's looks like quite a challenge to use digital
cameras to capture the shot flying past and then have usable data to
calculate positions.  I think I am going to try to use the paper scanning
method.  The paper we use already comes on a roll.  I will probably try to
use a 4 foot fluorescent bulb for a backlight to a giant fax machine
scanner.  Yes, once this process is sped up, we would like to use multiple
shots from the same combination to estimate variability.  And, for now, I am
going to ignore the 3-D pattern, just too much to deal with. I will get back
to you with more when this is out of the vaporware stage.
Thanks and more thanks.

Tony

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\23@210604 by Olin Lathrop

face picon face
> I would suggest that the only practical way of obtaining this information
is
> to record the shot passing through a light curtain with a high speed
camera,
> digitising the results, then comes the easy bit, mathematical modification
> to simulate the effect of the moving barrel.

Wouldn't you have a problem of multiple pellets in the light plane at a time
obsuring each other?


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, @spam@olinKILLspamspamembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\01\24@042441 by Philip Martin

flavicon
face
In actual fact, the correct way to use a pattern plate is by tracking the
gun across an arc and as you pass the aiming mark on the plate you fire.
This is supposed to give a more realist interpretation of the shot pattern
during actual use.

Tony,

How about fitting a sensor on the pattern plate and (using a PIC of course)
measure the time difference between the first hit and the last. This would
give you an indication of the length of the shot pattern, given that you
know roughly what the velocity is. Hell, put a sensor on the muzzle and we
could even check the velocity.

Hmm, I wonder how they'd feel down the club if I turned up with me gun and a
box full of electronics?

Philip Martin.




<big Snip>
In actual use the spread of the shot is also modified by the moving barrel
as you track the target. Therefore as well as knowing where the passes
through an XY plane perpendicular to the gun barrel you also need to know
when it passes through that plane.

I would suggest that the only practical way of obtaining this information is
to record the shot passing through a light curtain with a high speed camera,
digitising the results, then comes the easy bit, mathematical modification
to simulate the effect of the moving barrel.

Chris

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...