About RockBench


RockBench is an open access site where geotechnical data - data about the geometry and character of rock cuts, rock slopes, and mine rock faces - are shared to the benefit of the science and engineering community. A general introduction to the ideas of RockBench was provided in a journal article (in press as of June 2012) in Computers and Geosciences.


Site9_a_small_stripe


RockBench includes:


  • Datasets with locations, dates, collection parameters and site conditions
  • Reference information on the data acquisition methods used
  • Notes about the general field of remote geotechnical characterization (in the Blog)
  • Interesting references that you may find useful

The RockBench Vision:


RockBench is a common repository built around shared standards related to assessing rock masses using LiDAR and Photogrammetry. It has been established by Matt Lato, John Kemeny, Rob Harrap, and George Bevan with the core idea that we all benefit from meaningful comparisons of techniques on common datasets. It is our intent that RockBench be an open, and perhaps even ‘social’ site in the long term. All resources placed on the RockBench are for open access to anyone interested; all materials are under an open sharing license.

From the RockBench paper (submitted to Computers and Geosciences, March 2012):
"The purpose of this paper is to unite a community of researchers and developers working on advancing the field of rockmass evaluation through remote mapping techniques. The authors have independently learned that the development of a hypothesis into and algorithm is possible, but the results are isolated. There is no ability to measure improvement versus prior publications...The idea proposed in this paper, and executed through the RockBench website, is to enable researchers and practicing engineers in the field of remote imaging and engineering geology to test innovative algorithms for the mapping of geological features using open source data. This will facilitate the benchmarking of results, comparability of algorithms, and hopefully spur development in the field. The authors of this paper represent practitioners and developers; application specialists and theoretical scientists; lidar and photogrammetry experts; together we
combined our ideas to enable comparative development of data processing algorithms. The success of this idea lies within the community."

"The sites chosen by the authors for publication and distribution through the RockBench repository were selected based location, geological setting, physical setting, imaging technologies used. The data available on this website represent igneous, sedimentary, and metamorphic rocks. The sites will characterize natural terrain, blasted road cuts, open pit mines, and tunnels. Over time, and with the acceptance of this website by the imaging community, the number of sites will expand and the database will be extended."

Getting Involved in RockBench:


RockBench is only as good as the datasets provided, and it is fundamentally a community site. Our current architecture requires the administrator to add new materials, so please feel free to contact us and provide new content for addition. We will attribute all sources. Please note that RockBench is ‘open’ so datasets provided are freely accessible to all; we do not have a mechanism for restricted access nor is it our intent to have such a mechanism in the future.