LIDAR Magazine

From the Editor: Uncertainty is Uncertain

A 347Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE

Welcome to the SPAR 2013 edition of LiDAR Magazine. This year’s SPAR event is being held at a new venue– the Broadmoor in Colorado Springs, which is in you guessed it–Colorado, April 1518. It looks like a great location, but certainly not like Houston where many people can decide at the last minute to attend. The keynotes include Greg Bentley, founder and CEO of Bentley Systems and Michael Jones the Chief Technology Advocate at Google. These are certain to be highly informative.

There is a third keynote that may be the most interesting as it will be offered by Professor (em) Heinz Ruther, Principal Investigator of the Zamani Project. Professor Ruther has kindly provided LiDAR News with an article in this edition describing the work that he is doing in Africa to document and preserve their cultural heritage. It’s an amazing story. Kudos to all involved in this effort.

In addition to digital preservation this edition includes such topics as support for the relief effort after a destructive tornado in Mississippi, a look at a real-time, full motion FLASH LiDAR sensor and two articles on mobile LiDAR. We are always in need of interesting stories so please contact me if you have an idea. We want to help you to get the word out so the industry can benefit from the lessons learned.

Democratization was the topic of the last editorial. Michael Raphael’s recent LiDAR News eNewsletter article adds fuel to the democratization fire. When you stop to think about the possible long range impact that 3D printing can have on the consumer it is mind boggling. You won’t have to buy things, you will just print them. Add to this the concept of 3D printing of human organs and you run out of ways to describe the possibilities.

As Michael notes the dual announcements by the industry leader in 3D printers, MakerBot that they will soon be offering a low cost scanner as well as team with Autodesk to promote the use of photography as a consumer method of capturing 3D geometry just confirms that democratization of 3D is going to happen sooner rather than later.

In preparation for a presentation to the Texas DOT Standing Committee on Surveying I have been doing some thinking and research about accuracy of LiDAR and laser scanning measurements. I wrote a short blog where I noted that unfortunately in our industry today in most cases "Uncertainty is Uncertain" As I noted in the blog an individual who . is very concerned about this is proposing that we add a field to the ASTM E57 data exchange format to indicate the uncertainty of each point. That would certainly be a step in right direction, but I wonder if this is possible. It seems to me we have a couple of hurdles to overcome. The first is the black box nature of scanners. We simply do not have enough information from the manufacturers about the design of the sensor to independently asses the calibration of a scanner. Admittedly these are sophisticated instruments that the average person would have no experience in working on and would most certainly be likely to do more harm than good in trying to "adjust" but the same can be said for a , robotic total station or electronic distance measuring instruments (EDMIs).

The ASTM E57.02 subcommittee has been working for over 7 years to develop a set of standard test methods for assessing the accuracy claims of laser scanner manufacturers. It is a very difficult problem to solve. There are just too many variables and unknowns given the current lack of information and interest in supporting standards.

The Leica ScanStation P20 claims to have the world’s first "…onboard Check & Adjust routine that lets users perform an electronic field check and adjustment of the instrument to ensure an optimally performing scanner." This could be a step in the right direction. I am not aware of any user experience with this feature. If you are, please let us know.

A couple of years ago I had a conversation with Fred Persi who now has his own consulting business, but at the time worked for Quantapoint. In the early days they built and used their own scanners. When I asked him about calibration he explained that they had a test lab where he had used a laser tracker to determine the exact position of targets in the room. He would then routinely check the calibration of his scanners using the known position of the targets in the test lab.

I don’t recall the exact accuracy level which Fred said he was achieving, but this methodology makes sense to me in that it compares the instrument to a "known standard" that has an order of magnitude better level of uncertainty than the instrument being calibrated. Many laser scanning service providers are claiming that they can achieve the same level of accuracy as the control network. This may be possible under ideal conditions, but I don’t think I would want to put my stamp on it.

When the EDMI was first commercialized by HP in the mid-70’s the National Geodetic Survey (NGS) began supporting the creation of calibration baselines. We established one at the University of New Hampshire in the late 1970’s. From the NGS website it looks like they are still supporting over 300 of these baselines throughout the U.S.

I wonder why we did not follow a similar procedure with scanners as we did with EDMIs and total stations. I would assume it was because they were not intended to be used for control surveys, but the issue of uncertainty is becoming more important particularly in the mobile LiDAR arena where people are pushing accuracy to the limit and beyond.

What we need are standard calibration facilities for laser scanners and mobile LiDAR systems. Unfortunately, I don’t think this is practical for a number of reasons. Until the end users demand change it is going to be up to the people who use the technology to apply sound geomatics principles and established best practices to remove as much uncertainty from the finished products as is required for the specific application.

Regards,
Gene Roe, LS, PE, PhD Managing Editor & Co-Founder LiDAR Magazine

A 347Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE

Exit mobile version