12 replies to this topic

[UWPhotoTech]

I'm setting up a Subal D2X system for a research project. Reef sections will be photographed and then stitched into a Panoramic image. For this particular project it's important that there be no curvature to the field and subject size must be uniform from forground to background. I'm thinking of setting this system up with a 35/2.0 AF-D lens and a Subal FP-68 flat port, but I'd like to hear other thoughts and suggestions. For land based studies this photographer normally uses a 50/1.4 AF-D lens and wishes to get similar results underwater.

[UWphotoNewbie]

I think you are on the right track but do help your customer understand the differences between photography topside and below.

You might be able to eliminate distortion with a lens like that but lighting the scene will be very hard for anything sizable. This isn't a problem on land because you can always back up to get a wider view. Underwater you will either loose contrast (ambient lighting) or light (strobe) in a hurry.

So underwater photographers usually trade WA distortion for sharpness and color and just "get close". This means wide lenses (even fisheye) focused very close (increased subject size distortion) and dome ports.

This is all pretty standard but if its a newbie you should make sure the differences and limitations are well understood.

[james]

Hi Dave,

Will the photographer be using a T-head or a square frame for mounting the housing? This sets the camera up at a known distance from the target and can be used for x and y measurements too.

I would set the camera up with a 1 meter t-head and a lens with a field of view designed for a 1 meter square box. That would mean a 90 degree lens...hmm.

I'd set up the T-head with one SS200 on each end to light the scene.

Cheers
James

[Rocha]

I've used a similar setup to what James described. It was simple and small, one Nikonos V and a 35mm lens (I think). Light came from four Ikelite 50 (one connected to the camera, the others in slave mode). The setup worked very well. The main inconvenient was carrying the thing around! And it was only a Nikonos V, I imagine it will be a lot harder with any kind of housing.

Frankly, I don't see the need for a D2x to do this kind of job. If I had to do it, even though I have a D2x, I wouldn't use it. If I had to use a PVC square like the one James described, I would just use a point and shoot. They work very well for tasks like estimating coral cover, and are a lot easier to carry around to make many photos along a transect line.

[james]

I'm moving this to the Scientific Photo Forum. I know Dr. Wood has done similar work.

Cheers
James

[UWPhotoTech]

First, thanks to each of you for the replies.

James, although I recall this being discussed previously I can't find any posts (other than this one) that discuss T-Head's or Square Frame's. Can you point me at some more information or other posts so I can make a recommendation?

This photographer is new to the underwater world but is well known above. To control lighting and perspective on land he has used 100's of images to make up his panoramic shots.

The camera and housing choice was made by the photographer.

[Paul Kay]

The 35mm lens/flat port sound pretty good to me. I always think that the 35mm lens is very underrated for use with a housed camera although I used to use the 35mm f/2 AF lens with the FP-68B port myself and found it very useful for certain subjects. However, using a 35mm lens and a flat port will result is some chromatic problems at the corners as the port is of 'significant' thickness. You may well want to sort out and stick with a fixed set of parameters for the Raw conversion, correcting this as appropriate. I suspect that some degree of distortion may occur too - I only used my set-up pictorially - again a fixed set of parameters in appropriate software to correct this may be of help.

I would not try to use a wide-angle and dome as field curvature of the image will almost certainly induce significant distortions which will probably require a good deal of correction, and corner softness may be an issue too. Neither will help if you are going to stitch images together!

If images are of a none moving subject such as a reef, it may be worth looking at lighting by a continuous light source (Kowalski HID lights spring to mind) as lighting could then be assessed as being even visually (just a thought).

For mounting you could try Really Right Stuff's Panorama Head (with the advantage of a quick release) which allows for easy rotation (how long it would last in salt water I don't know but it is hard anodised.

[UWPhotoTech]

Thanks for your thoughts Paul, it seems like we've come to similar conclusions. I was aware of the potential for chromatic aberation problems but felt they would be minimal with the 35mm lens as it's effectively a 52.5mm lens on the D2X. I ruled out a dome port for the same reasons you've mentioned.

I will pass on your comments on the image processing, lighting and of the RRS panorama tripod head. Which by coincidence is on my personal shopping list at the moment.

[Tom_Kline]

The Hasselblad booklet on UWP showed something like what James' described. These are sold frequently on the bay. Be aware that there will be some pincushion distortion with a flat port but it may an acceptable amount given that the lens probably has some distortion too. It would be smart to do tests in a tiled pool. Assess the test pix for distortion using the squares. B)
Tom

[UWPhotoTech]

After getting a search recommendation from James I found a decent photo of the square frame setup, here's a link to the page:

http://www.hawaii.ed..._monitoring.htm

[Rocha]

Dave, with digital there is not much need to attach the camera to the square frame. We did that on film just to make sure that we were photographing the entire quadrat. With digital's instant feedback you can check if you got it or not immediately as well as check the alignment. By eliminating the sides, the resulting system will be a lot less bulky and easier to carry around since it will consist of only the unnatached housing and the square.

[james]

That's why a T-head is better. Camera mounts at the apex of the tee with a strobe at the end of each arm.

James

[UWPhotoTech]

Thanks Luiz and James! My part in this process is the lens/port recommendation which I think we've established. I've passed on a link to this thread to the photographer so he can read your comments on the procedure and tools.