Jump to content

A Forest Clearing Near My Home in Sydney


tony_dummett

35-105mm zoom lens, ISO set to "180", 36 (3 rows x 12 columns) panels stitched as a mosaic with Panavue software. 4 sec. exposure @ f20 per panel, using a special "contraption" of my own design to eliminate three-dimensional parallax errors.

 

See my comment below for comments on creating mosaics.


From the category:

Uncategorized

· 3,406,231 images
  • 3,406,231 images
  • 1,025,779 image comments


User Feedback

Recommended Comments

This is a detail of the above picture (at 50% resolution, taken from just to the right of the small, central tree, mid way vertically).

 

The aim here was to see if I could stitch a mosaic of pictures together to make a large, high-definition whole.

 

To do this I needed to construct a contraption that screwed onto my panoramic head to allow three-dimensional panning and tilting of the camera, while keeping the "nodal point" of the lens (its optical centre) in exactly the same position, irrespective of the attitude of the camera. Geometrically, the nodal point of the lens was to be the point of a cone, with all pictures - looking up, down, left, right - taken from that position. If the nodal point is aligned this way, there will be minimum parallax errors between objects in the picture.

 

Taken to its ultimate, the nodal point of the lens really needs to be the centre of a sphere. The ultimate extension of this technique is to allow the construction of spherical panoramas (this is really only a practical proposition if you have use of a suitable wide-angle - even fish-eye - lens and a trick tripod that doesn't get in the way when you tilt the camera 90 degrees down).

 

Other examples I've seen arond on the net use Professor Dersch's PanoTools software, but I didn't have a year to spare fiddling around with PanoTools - which, despite attempts by third parties, does not have an intuitive GUI yet - to make this mosaic stitch a reality. PanaVue was fine, but there's a trick to it (hint: use cylindrical, NOT spherical, projections for your initial stitches).

 

I included the small detail upload above to show how much you can capture using this method. The detail is a 50% resolution extract from the final, full-sized stitch. However, the "full-sized" stitch is itself a reduction to 75% size from the camera's native (100%) resolution. I guess that means the detail shown is 37.5% of the original camera's resolution (50% x 75%), or a little over one-third native resolution of each of the 36 pictures I took down in among the trees.

 

A word of caution: stitching 36 panels together, each of 17 megabytes, was just too much for my (otherwise really good) Panavue stitching software and I kept getting "memory errors". At full camera resolution, the final picture would have been over 540 megabytes, and if rolled out on a suitable flat surface, could have been used as rainproofing mat for a tennis court, or perhaps a roof for the Astro Dome (a doona for Gulliver? backdrop for a G. W. Bush campaign speech?... you get the point).

 

I also showed the detail to make the point (hopefully successfully) that this picture works best in great enlargement, not as a thumbnail of the original (which this Photo.Net version truly is). That it works better in one enlargement than another led me to ponder whether there exists what could be called a "fractal" aspect to a photograph.

 

In fractal mathematics each level of observation - "zooming" in or out - is just another iteration of a simple algorithm. The detail (or perhaps "beauty") of the system is the same at every level. Does a good photograph have to "work" for the viewer in the same way? Can a photograph be good at high resolution but less good (or even bad) at a lower resolution?

 

Or are the best photographs (and any other artworks) good at every level? Should they be like a fractal system? Or can good photographs have different appeal at different levels? That would make them similar, but more complex in appeal to a fractal system.

 

 

870116.jpg
Link to comment

Interesting, what's next? May I suggest a macro lens mounted to a twelve foot magnesium concrete striker? I'm thinking of a mosaic of about 144 columns wide by 72 rows high at 1/1 magnification. That should keep you busy for a few hours.

 

It's good to see that your new toy has not led to any sort of weird or dangerous obsessions. I would not like to travel all the way down there for some kind of an intervention. I had better not hear any reports of a strange, filthy man wearing only an old burlap sack, lugging camera equipment through the woods and talking to himself about the spiritual nature of jigsaw puzzles.

 

Seriously, thanks for sharing the technique along with the photo (like I could be serious). By the way, you and the dog are starring in a new POW tribute.

Link to comment

Tony

 

Cracking pano' technique. Love to see it big! How are you going to print this mother though??

 

You get a 7 from me for originality due to the unique contraption you made.

 

Cheers

 

 

Link to comment

Printing Big?

 

My local (well, not so local) printing emporium, had trouble with my last Monster... at a mere 340 meg.

 

I got all kinds of, "So sorry", "Firl [sic] too big", "Come back nex [sic] time", kinda remarks that I allowed myself to become depressed. But I've recovered now. I'm going to force my will upon these lab people... Bigger IS Better!

 

That's how we do it in Ozzie. Big and Bold.

 

The "contraption" is really just a piece of 3mm aluminium (for my U.S. readers: "aluminum"... I mean if GDubyaB can say, "Nukular", why can't y'all say "aluminum"?) - and by the way you were close Dennis, "Al" is just one removed from "Mg" on the Periodic Table - with holes drilled at "strategic" places. I really need a slot milling machine, but didn't have one (neither do I have the space under the house for one... thanks to YOU KNOW WHO my-next-door-neighbour, Dennis), and didn't want to pay for one until I knew the technique worked. Now that I DO know it works, what's my excuse?

 

None. OK, I'll get it printed.

 

As to "dangerous obsessions", D., surely you know me better than that?

 

 

Link to comment

I'm amazed at the way you always find something new every time you publish a picture here.

 

The picture, ok, is an effective pretext to pre-test your wits. But, what about the planarity of your CCD? isn't it a bug in the entire spherical affair? :)

A shape-controlled elastomeric substrate for the CCD is an idea: I'll work on it.

ciao,

giacomo

Link to comment

G'day from Down Under! (That's Tassie compared to your location). Giacomo's comment made not much sense to me at all, so I guess he is on a different technical level than me...and so are you. Just an honest (ignorant) observation, to the average punter, your final image doesn't look like a super-panoramic made up of 36 individual frames. So, I hope that some will stop to read all the technical details and appreciate the amount of work that went into this (eventhough that is not supposed to influence our evaluation of the final result). Oh, you are wondering about this comment heading...

I was brought up on Canon equipment and just finished writing an extensive email tonight about people that suffer from "brand loyalty" (myself included). So, the question is, what made you go for the Nikon D100 as opposed to some of us that would go out and have a bash with the Canon 10D, instead? I realise that this is going off on a tangent and taking the discussion away from your "Forest Clearing", but I was just curious... Cheers,

Link to comment
Yes Peter ! This would be interesting for me too! Up to now I'm used to Canon... but now I had a Leica R8 and some fine glasses to play arround, and suddenly the Canon looks/feels like a toy - I'm not sure if the images will be that much better - let's see, but the handling is that MUCH different - so that I will go out tomorrow and try some Nikon stuff, just to see where they are...
Link to comment

Yeah Tony, why the D100 and not the 10D? As one who is looking to get a fully fledged DSLR in the next few months, i need to know and pray you'll tell me, what swung you to the Nikon?

 

Cheers

Link to comment
Why Nikon and not Canon?

I had decided to sell my Pentax 67II (and any of the related lenses, filters etc. that were saleable along with it). Why? Well, it's not really relevant to this discussion, but I'll tell you anyway: I wasn't using it. I wasn't using it because of the hassle and expense of film: buying, processing and scanning. And I thought I'd better try this digital business out.

Being constrained by a temporary impecuniosity (due to the inordinately expensive and endlessly finickerty renovation of our other house in Sydney, prior to sale), I didn't have the readies to justify both the Pentax and a digital device.

Phil and Bruce, the redoubtable proprietors of The European Camera Specialist store in Drummoyne Sydney, has a "special" at the moment on D100s. Not much of a special, really, but Chris Battey rang me and told me about it (and had whetted my appetite a few weeks before by lending me his D1 for a few days).I could either have the D100 immediately on a trade-in basis (plus a few hundred in cash), or selll the Pentax on consignment, take the cash elsewhere and buy something else (or perhaps even spend the receipts of the sale on more renovations). Bruce, never one to miss a deal when he sees a mug pitching him one, made me an offer I couldn't refuse for the Pentax gear and I walked out with the D100 (plus a 24mm 2nd hand lens in superb condition).

If I hadn't been so bull-at-a-gate about "going digital" immediately (yes, it was an impulse buy), I might have waited a little while and seen the price drop along with the U.S. dollar. If I had done so, then the extra cash injection to make up the total sale price might not have been needed. But when you impulse-buy, you don't care about things like that.

It's a fine camera and does most of the things I'd like it to do. It's a good entry-level device, and for my stitching activities is as good as any other. At least, that's what I've told myself, and now I'm starting to believe it.

Sorry my explanation isn't very profound.

As to aesthetic considerations: the fact that this pic is the result of many days thinking and planning to get to the point where I could capture and stitch the panels seamlessly, is entirely irrelevant to whether it is any good... or not.Now that the process is understood by me, I hope to make a few more spectacular shots than this in the coming little while. As I said above, it's worth is in its detail, its "micro" aspect, rather than its "macro" aspect. I may get it printed just to see whether it looks spectacular enough to frame.

Link to comment

Tony, I love these woods pictures. I can't do them nearly as well, of course, but I keep walking out into the woods, setting up my tripod, and snapping single frames that I don't stitch together(my folder: "Before the Europeans"). Mine are mediocre for any number of reasons.

 

I think that you have the right idea: it does have to be big, and it does have to have incredible resolution, to capture the essence of the vastness and richness of the woods. Even so, resized to 800x600 (or whatever), the resolution still comes through--not like it does with the tennis court sized version, but it is there. The quality shows, that is, even at this size and after upload. If I could only see what you are seeing full-sized.

 

Another issue is the subject. "What is your subject?" they ask. Why, the subject is the woods, of course!

Link to comment

I've been missing from photo.net for some months (just took a peek sometimes to the POW but I must say that there were just a few people who made interesting comments - one was Tony - and I noted that most of them don't make comments any longer, so definitely I don't care POWs anymore).

 

Then I was curious if there were news in Tony's folders. It's always a pleasure to see Tony's inventive. BTW, I've just bought a D100 too. And the idea of these mega-stitches is intriguing (obsessive? Maybe, but fun!). I won't be as skilled as Tony, but I'll give a try.

 

See you later!

Link to comment
Incidentally, there was an article in Scientific American some months back analyzing the fractal character of Jackson Pollack paintings. Apparently, the authors were able to tell the genuine works apart from imitations by mathematical analysis.
Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...