Oct 292012

For those CDAA members who attended the talks associated with our Annual General Meeting over the weekend, you will remember JDZ’s introduction to some new cave mapping techniques. Cave explorers are familiar with counting knots in the line and noting depth, direction and azimuth all the way home. Dive complete, you exit to dry land, translate your scribbled notes for entry into a computer program and check out the stick map of your progress. This time consuming exercise has been the only way for your average diver on a budget to produce accurate maps of our underwater travels.

Beyond being time consuming, errors can creep into a survey over time. This is especially true in complex caves with multiple branching passages. Ken Smith has become justifiably famous for his pinger technology, which allows points underground to be pinpointed in the world above. Being able to ping key locations and use those co-ordinates to align known surface features and the underground map is fantastic for checking the mapping work while it’s underway.

And now it appears that underwater mapping is about to make another giant leap forward. There are several options currently being explored. The two that JDZ mentioned in his talk were combining multiple photos of a cave into a 3D mesh field, and using two closely spaced and synchronised video cameras to generate a 3D model of the cave. While the dual video method holds more promise of one day producing immediate results, we haven’t yet had success in preliminary testing.

On the other hand, the photographic method has produced a number of cave models with programming pioneered by Wayne Johnson. After the talks and AGM on Saturday, JDZ and I went for a quick dip in Tank Cave on Sunday. Between our artistic efforts, I paused to take a sequence of 38 shots of this pillar in D tunnel. On the drive home, I processed the shots and put them through the 3D engine. In less than 10 minutes we had a spatially accurate 3D computer model of the cave area – absolutely amazing.

Click through to here to see the model we generated for yourself. Depending on your browser (Chrome worked for me) you should be able to load the 3D model in a viewer without installing any programs. Check it out, have a play, and imagine the possibilities!

3D model of the pillar

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>


(required, will not be displayed)