A 923Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE
I really need to clean my office. A quick count yields 16 hard drives with over 35 terabytes of LiDAR. Although I sometimes feel overwhelmed by all the data I have, and resent the fact that my job sometimes resembles that of a UPS store clerk rather than a remote sensing analyst, I have it relatively easy. All the people that need the data on those drives are practically within arms reach.
What do you do if you are part of an organization working on a project involving large LiDAR collections, with team members scattered across the country, all with different needs? You have a remote sensing technician in Portland who wants all the LAS files in their original projection; you have a cartographic production team in Des Moines that wants contours for an area of interest in the respective state plane coordinate system; and your engineering team in Atlanta wants raster surface models for only a small transportation corridor in a geographic coordinate system. Do you generate products that will meet every possible requirement and send out multiple hard drives? Do you purchase expensive software for each group and train them to process the data themselves? Whatever route you take it is going to cost you time and money.
What if there was a cloud-based solution that allowed you to house all your LiDAR data? What if that same service allowed your user base to easily find and locate the data for their area of interest? What if it allowed them to not only download the LAS files, but also generate derivative products such as raster surface models and contours, and in the coordinate system of their choosing? That would make your life a heck of a lot easier, wouldn’t it? You wouldn’t need to spend time generating various products in host of coordinate systems and figuring out ways to get said datasets to your users in a timely manner.
Fortunately you are not the only one thinking this would be a great idea. The team at i-cubed (i3.com) has developed a solution and it’s called DataDoors. DataDoors was originally developed as an in-house solution to organize and share data, supporting i-cubed’s early work with satellite imagery. It didn’t take them too long to realize they had a unique solution and the Army Geospatial Center (AGC) signed on as the first DataDoors customer. DataDoors quickly expanded into the oil and gas sector, which like the AGC, were large organizations with a distributed user base that needed access to large amounts of imagery. DataDoors is also an e-Commerce site used by the major satellite imaging companies to sell their data.
Most recently the i-cubed team added LiDAR capabilities, and not just the capability to store and distribute static LiDAR products, but to generate derivative products. I decided to take DataDoors for a spin and the i-cubed folks were kind enough to load a small hard drive of imagery and LiDAR I sent them onto the DataDoors server. DataDoors can run on your own infrastructure, but most organizations choose to have i-cubed host the data for them as i-cubed has put a lot of effort into insuring the performance and redundancy of its systems.
DataDoors can be accessed at www. datadoors.net. The entire solution is browser-based and does require the Silverlight plug-in. Logging in is not a requirement, as guest users can browse datasets open to the public, but it is a requirement to have an account if you want to order data. Organizations seeking to use DataDoors will appreciate the fact that access to data can be tightly controlled with permissions settings available at either the group or individual level. If you have sensitive or proprietary data this insures controlled access to your data. The interface for DataDoors is clean and easy to navigate, even for the first time user.
The main functionality of DataDoors is to find and order data, but it also offers the ability to annotate the map and create projects that you can share with other members of your team. I won’t be covering the last two features in this review, but such capabilities are a definite asset for those who work in collaborative environments. DataDoors makes it easy for the end user to find data, termed "discovery." You can search by place name, coordinates, county, quadrangle, PLSS, a user-defined point, a user-defined rectangle, a user-defined polygon, or you upload a shapefile. It seems as though the folks at i-cubed have thought of every possible way to discover data.
Once you have established your area of interest DataDoors queries the data holdings for which you have permissions and returns the search results, all in a matter of seconds. I wish my desktop GIS software could do spatial queries this quickly. You can filter the results of the query based on the resolution, data, cloud cover, and product. My favorite feature is that a slightly reduced resolution version of the product is displayed on the map interface whenever you mouse over a dataset in the search results. Pretty slick and far better than simply displaying a wireframe. For LiDAR datasets the displayed product is a colorized digital surface model.
You can individually, or in batch, add datasets to your shopping basket. Once the datasets are added to your basket the reduced resolution image is added to the map. This makes it extremely easy for the end user to ensure they have both the correct products and that they have complete coverage for their area of interest. From the basket it’s one-click to the checkout.
The checkout is where a lot of the magic happens, particularly with respect to LiDAR. You can choose to download the original LAS files or you can derive contours, a Digital Elevation Model (DEM), a Digital Surface Model (DSM), or an intensity image. For the raster products you can change the format (e.g. GeoTIFF or IMAGINE) and for all products you can select the coordinate system. You can also create seamless raster mosaics from multiple LAS files.
Once you have completed the ordering process you will receive an email with download instructions. It should be noted that when it comes to deriving information, whether it be contours or surface models, from large LiDAR datasets the processing is going to take some time, even on a local workstation. My experience was that LAS files were delivered within a few minutes, while derivative LiDAR products took minutes to hours depending on the size.
In the LiDAR community we spend a lot of time talking about new technologies such as sensors and automated feature extraction, but we typically ignore one of the great challenges, which is data distribution. DataDoors offers a compelling solution for distributing and processing remotely sensed data, whether for internal use or for sale. It is ideal for organizations that have personnel that are geographically separated. The cloud-based processing environment allows organizations to publish LAS files, but give their user base access to multiple types of derived products in an environment that is suitable for novice users.
Jarlath O’Neil-Dunne is the Director of the University of Vermont Spatial Analysis Laboratory. His research focuses on developing automated techniques for extracting information from high-resolution remotely sensed data.
A 923Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE