Underwater RTIs?

[Dennis Piechota]

Does anyone know of attempts to apply H- or D-RTI methods to underwater archaeological sites?

It's a tall order but there is currently a move to leave artifacts on the seabed (with no or little artifact retrieval to land) and create video links between coastal museums and the sites for research, education and monitoring. D-RTI imaging of artifacts in situ would add greatly to that experience and to our ability to interpret finds and their matrix. Objects viewed underwater look completely different from land. With specular imaging and other rendering modes those artifacts and the seabed surrounding them will be amazing. 

Dennis

Dennis Piechota

Archaeological Conservation and Micromorphology

Fiske Center for Archaeological Research
UMass Boston
Office: 617-287-6829

XRF Group Admin

pXRF for Cultural Heritage

RTI Group Admin

RTI_help

[Earl G.P.]

There was someone who spoke at our Oxford workshop a couple of years ago who did it. I believe we have also done pool trials here in collaboration with Centre for Maritime Archaeology. Hembo will know if it was ever sorted.

Cheers

Graeme

Office: 617-287-6829<tel:617-287-6829>

XRF Group Admin
pXRF for Cultural Heritage<http://groups.google.com/group/pxrf>
RTI Group Admin
RTI_help<http://groups.google.com/group/rti_help/>

[Moshe Caine]

No, I don't. However the very thought of underwater RTI is mind blowing. I'd love to see results.

Sent from my iPad

[Tom]

I've been approached over the years several times about applying PTMs underwater, I think the idea has spectacular consequences. In my opinion doing the imaging and reflectance transformations in real time would be the way to go. The 13th paper down on my home page, http://www.hpl.hp.com/personal/Tom_Malzbender/, entitled "Surface Enhancement Using Real-Time Photometric Stereo and Reflectance Transformation" shows that it can be done at video rates using a GPU. One could imaging driving an ROV around and seeing enhanced imagery at video rates from the sea floor. I am hopeful it would also help with cloudy water. Since the distribution of particulates should differ for varying light sources, they should get cancelled out in the final images to some degree. I would welcome any collaboration on this, this should have deep impact on underwater archeology and is doable with todays technology.

 

Tom Malzbender

[Ryan Baumann]

Two things that might also be helpful to keep in mind are the light
scattering effects of water as well as its selective wavelength
absorption (i.e. "green" will travel further - you may in fact want to
exploit this). Depending on the object and conditions, I imagine
scattering may even cause secondary undesirable artifacts at hard
shadow edges. A starting point for trying to overcome scattering
artifacts might be
http://www.cs.cmu.edu/~ILIM/projects/LT/LightTransport/lighttransindex.html
& http://www.cs.cmu.edu/~ILIM/projects/LT/structured/structured.html
and work that cites it.

-Ryan

[Dennis Piechota]

Great Tom! I'm actually leaving in a couple hours to take part in a deepwater ROV survey in the Black Sea. A small part of ongoing work that is being webcast 24/7 for the next 2-3(?) months. You can see and participate (ask questions) in the activities at: http://www.nautiluslive.org/

I'll be on board from the 19th to 24th to carry out artifact decay monitoring. Besides the underwater museum applications I'm interested in using RTI imaging for periodic re-evaluations of surface alteration to decaying artifacts in situ. 

Enhanced rendering of objects on the seafloor at video rates is mind blowing indeed! Just 'canceling out' the marine snow alone as you suggest would improve our image quality greatly.

Dennis

Dennis Piechota

Archaeological Conservation and Micromorphology

Fiske Center for Archaeological Research
UMass Boston
Office: 617-287-6829

XRF Group Admin

pXRF for Cultural Heritage

RTI Group Admin

RTI_help

On Mon, Jul 16, 2012 at 12:26 PM, Tom <tom.mal...@gmail.com> wrote:

[Earl G.P.]

This would indeed be fab! In addition to pool testing we also spoke to some ROV people here at Southampton – again I don’t think it has taken off yet but I would also welcome picking it up. Cheers,

 

Graeme

[Dennis Piechota]

In my chats with the ROV engineers on this cruise they are as always willing to make anything happen. They point out though that adding any new functionality to an ROV toolset limits or compromises the other components of the system. So consideration of how an RTI capture and lighting system can 'play well' with the rest of the streaming data and with the trim and buoyancy of the vehicle is critical. It's easiest to make a system that requires re-fitting for dedicated dives but that would limit the application of RTIs to special ops. 

Here are some considerations from the point of view of our ROV system (Hercules) as I understand it: 

We do not take still images; we do high-definition video only and from that we grab TIFFs as stills (6.2Mb @ 1920 x 1080 pxl each). An RTI system would either have to use that or have a separate camera and communications hub to join the stream of imaging to the control van for RTI processing. Tom's mention of high capture rate RTIs may work well as a direction to develop towards given our current digital video strategy. 

But to get off the ground first, and since we and other ROVs use two manipulator arms they could be used in the beginning to do 'proof-of-concept' H-RTIs (one arm lit while the other holds the spheres) with the intent of demonstrating the potential of a large dome lighting/capture system.

We have recently moved away from custom color correction at the seafloor because of our streaming video (hard to use clips where the color balance keeps shifting). In a way this 'raw' look would make rendered RTIs all the more appealing if the processing can be optimized for speed to make their production close enough to real-time so as to be compared with the live imaging.

In all this I remind everyone that I am not an engineer but a conservator/archaeologist who stands on the shoulders of others and appreciates the worlds of both engineers and scholars. An enjoyable place.

Dennis

Dennis Piechota

Archaeological Conservation and Micromorphology

Fiske Center for Archaeological Research
UMass Boston
Office: 617-287-6829

XRF Group Admin

pXRF for Cultural Heritage

RTI Group Admin

RTI_help

[Hembo Pagi]

Hello there

We prepared all the gear here but sadly we still have not had a chance to do the pool test. Hopefully we do it this autumn.

All the best

Hembo