"Trevor G Quayle" <Tin### [at] hotmailcom> wrote:
> 3) This is a little secret I didn't want to describe yet, as it can be tricky to
> master and involves a little math:
[snip]
> As I said, this can be tricky and tedious, but I find it does help somewhat if
> you can get it right.

Cunning indeed. I may give this a try if all else fails.

It sounds like distant environments centred properly in shot will be the way to
go for the time being, however.

Thanks for all the detail!

Post a reply to this message

"Trevor G Quayle" <Tin### [at] hotmailcom> wrote:
> > > Maybe post some details of your difficulties, or if you'd like you can email
> > > me directly.

Here's a pair of angular maps that demonstrate my issue (scaled down to 512x512,
but it's still evident). I've marked corresponding pixel locations with red
crosses.

http://www.infradead.org/~wmp/angular_eg.png

It's interesting, now I come to actually look at this in detail, it seems to be
most pronounced on the horizontal. This suggests to me that keeping the ball
more central will definitely help...

Bill

Post a reply to this message

"Bill Pragnell" <bil### [at] hotmailcom> wrote:

> That's about right, my camera's 7.2Mp and I can juuuust get it focused at full
> zoom from about half a meter away, although I can't get it to focus full-frame.
> I'm getting 1700 pixels square after cropping. Perhaps I'll try the focus
> override - forgot about that.

Oh - one more thing I forgot; if your camera has an actual aperture (and many of
the Canon point an shoots don't) try to set it to the smallest value to increase
the depth of field.

> I think it's stainless steel, so I shouldn't have any problems. I've never had
> stainless steel rust before except through wet contact over months. Where on
> earth can I get it chrome-plated, and how much would that cost?! :)

I thought mine were stainless steel, but maybe they were just steel...

Chrome plating is usually a function of how big the object is - and a ball
bearing is pretty small. There's often a minimum charge involved though. Mine
was about 20 pounds because of that minimum. If I were to do more balls, I'd
shop around to find if someone could do it cheaper. Just look in the yellow
pages for chrome platers in your area.

The only tricky part of how to mount the ball in the chrome plating tank. I got
them to weld a wire onto the ball and hang it into the tank. You end up with a
spot that's, but that can go on the bottom. I also got some small bubbles on
the opposite side, but there's enough clear area to take a good picture.

> > I use HDR shop to convert the images from spherical mirror projection (i.e. the
> > HDR photograph) into Latitude/Longitude format, then do the stitching in
> > Photoshop. I think there is an HDR version of GIMP - Cinepaint? Everything is
> > much simpler to do in square lat/long format, and POV can use the resulting
> > images just fine.
>
> Actually the blending is very straightforward in HDRShop, I can knock up a mask
> in the GIMP in about 5 minutes. I don't have Photoshop, so I am limited in my
> retouching facilities. I should see if Cinepaint works under wine...

Using Photoshop (or something like Cinepaint) is just a preference I have, but I
do find it easier to manually touch stuff up in lat/long format.

Cheers,
Edouard.

Post a reply to this message

"Bill Pragnell" <bil### [at] hotmailcom> wrote:
> "Edouard" <pov### [at] edouardinfo> wrote:
> > One of my shots - to give you an idea of the source material I'm getting with a
> > 1 inch ball.
>
> That's great - much clearer than I've managed so far even with my bigger ball. I
> really need to polish it and investigate that focus override... :)

Thank-you. It took a long time before I could control everything enough to get
good results. The in-focus plane is very narrow, and that was the biggest
problem I was facing.

The CHDK grid I used to help me (you'll have to adjust the circle to you zoom
and ball size):

@title LProbe
@elps   180, 120, 46, 41, 0x01
@line 179,   0, 179, 239, 0x07
@line   0, 119, 359, 119, 0x07

Cheers,
Edouard.

Post a reply to this message

clipka wrote:

> The more zoom you use, the less prominent these distortions should be.

I wonder if anyone ever tried to do an outdoor mirror ball shot
using a backyard telescope with ccd from really long distances ;)

Post a reply to this message

"Bill Pragnell" <bil### [at] hotmailcom> wrote:
> "Trevor G Quayle" <Tin### [at] hotmailcom> wrote:
> > > > Maybe post some details of your difficulties, or if you'd like you can email
> > > > me directly.
>
> Here's a pair of angular maps that demonstrate my issue (scaled down to 512x512,
> but it's still evident). I've marked corresponding pixel locations with red
> crosses.
>
> http://www.infradead.org/~wmp/angular_eg.png
>
> It's interesting, now I come to actually look at this in detail, it seems to be
> most pronounced on the horizontal. This suggests to me that keeping the ball
> more central will definitely help...
>
> Bill

Actually, by the looks of it, I think that the technique I described may help
for this as there does appear to be lopsided distortion in the rotated image,
as I would expect from perspective distortion.  Give it a try, let me know if
you need any help getting it to work.
What is your ball size to distance ratio?

-tgq

Post a reply to this message

Christian Froeschlin wrote:

> clipka wrote:
> 
>> The more zoom you use, the less prominent these distortions should be.
> 
> I wonder if anyone ever tried to do an outdoor mirror ball shot
> using a backyard telescope with ccd from really long distances ;)

If you're going for long distance then why not use Hubble? :)
-- 
Ger

Post a reply to this message

"Trevor G Quayle" <Tin### [at] hotmailcom> wrote:
> "Bill Pragnell" <bil### [at] hotmailcom> wrote:
> > "Trevor G Quayle" <Tin### [at] hotmailcom> wrote:
> > > > > Maybe post some details of your difficulties, or if you'd like you can email
> > > > > me directly.
> >
> > Here's a pair of angular maps that demonstrate my issue (scaled down to 512x512,
> > but it's still evident). I've marked corresponding pixel locations with red
> > crosses.
> >
> > http://www.infradead.org/~wmp/angular_eg.png
> >
> > It's interesting, now I come to actually look at this in detail, it seems to be
> > most pronounced on the horizontal. This suggests to me that keeping the ball
> > more central will definitely help...
> >
> > Bill
>
> Actually, by the looks of it, I think that the technique I described may help
> for this as there does appear to be lopsided distortion in the rotated image,
> as I would expect from perspective distortion.  Give it a try, let me know if
> you need any help getting it to work.
> What is your ball size to distance ratio?
>
> -tgq

Have a look.  I attempted to deconstruct, unwrap, unrotate your images and apply
the technique I mentioned.  Its not perfect (and I didn't have the original
unrotated image to work with) but the distortion is far less noticeable.  I
added about ~4% to the angular map (+20 pixels to each side of a 1024X1024
image)

-tgq

Post a reply to this message

Attachments:
Download 'newprobe.png' (689 KB)

Preview of image 'newprobe.png'

Just to beat the topic to death...

This is another instance where Lat/Long mapping is helpful.  If you look at the
original and corrected lat/long versions below, you can clearly see in the
original one how everything 'pinches' in to the point.  This basically
indicates that it wasn't a true 360 mapping.  In the corrected version, the
pinching is far less noticeable (however there still is a little bit if you'll
notice the doorframe behind, which tells me I didn't add quite enough to the
map, maybe I should've done 25 each side instead of 20)

-tgq

Post a reply to this message

Attachments:
Download 'lb1.png' (1422 KB)

Preview of image 'lb1.png'

"Trevor G Quayle" <Tin### [at] hotmailcom> wrote:
> Just to beat the topic to death...

Wow, that really makes a big difference. I'll try this on the images I've taken
so far from scratch. Thanks for taking the time over this, it will be a massive
help! I was aware the ball is a few degrees short of a full 360, but knowing for
certain that this is the cause of my problem is most of the work.

(haha, looks like I have no legs from the knees down - relax folks, I'm kneeling
;-)

Post a reply to this message