airbornelasermapping.com

A reference source on an emerging technology

December 1998:     PacNet Reporter interview with Robert Eadie of EagleScan Inc.

In December Robert Eadie, Manager, Marketing and Sales and the rest of gang from EagleScan invited the intrepid PacNet Reporter out to Boulder, Colorado to get in some virtual extreme downhill, sip some egg nog around their tree and discuss the DATIS™ system.

PR:    Can you give us a little background about EagleScan? Oh, and perhaps put the egg nog over here besides me?

RE:    Sure. - (g) - Established in 1994, EagleScan has developed laser-based mapping solutions that offer new, higher standards in accuracy, affordability, and timeliness when compared to traditional topographic mapping methods. These advantages are accomplished by acquiring topographic data using an airborne sensor package known as DATIS™ (Digital Airborne Topographical Imaging System™). We use DATIS™ to collect not only high-accuracy terrain elevation data or Digital Elevation Models (DEMs), but also complementary images of the terrain. Our products include high density and high-resolution digital elevation models (AccuDEM™) and digital panchromatic images (EagleImage™). The DEMs are created from a high density grid of elevation samples, several orders of magnitude denser than standard photogrammetric approaches normally provide. The complementary imagery is collected using a large format digital camera, which acquires panchromatic images with appropriate overlap. A key feature of this process is timeliness. We are able to collect, process, and deliver these products in days or weeks as compared to weeks or months for competing approaches. More details are available on our web site.

PR:    Can you give us some technical details on the DATIS™ system?

RE:    Well, the DATIS™ hardware itself integrates a LIght Detection And Ranging (LIDAR) instrument to measure terrain elevation using laser range finding. A scanning mirror is used to direct laser light to and from the laser transceiver. A relative position GPS (Global Positioning System) is used to measure the precise aircraft position and an Inertial Measurement Unit (IMU) is used to measure the aircraft attitude or pitch, roll, and yaw. There is also a digital camera on board DATIS™ that is mounted to the same optical platform as the LIDAR. This is used to collect corresponding digital panchromatic images that may be orthorectified and thus georeferenced to the DEM data. During a survey DATIS™ directly measures the topography of the earth's surface to capture topographic and geographic information unachievable with traditional survey methods. This system also provides capabilities to generate other products such as planimetric feature maps, three dimensional high-resolution image models and precision digital ortho rectified images to support various GIS needs.

PR:    And what are some of the benefits of using DATIS™ to conduct field surveys?

RE:    A few of the benefits include:

• Up to 220,000 elevation samples per square mile results in precise Digital Elevation Models (DEMs).
• Fully automated digital topographic data collection and processing results in fast data delivery and lower costs to the customer.
• Variable modes of operation and spatial data resolutions to facilitate a wide range of mapping application needs.
• High-resolution digital panchromatic image data collected simultaneously, which can be orthorectified to provide planimetric feature information.
• Only one ground control point is required for most projects reducing costs and time compared to photogrammetric methods.
• Data can be collected at any time of the year and even at night in some cases increasing the likelihood of fast turnaround and delivery of map data.
• Widest LIDAR swath width in industry allows us to collect up to 75 square miles of elevation and image data per hour. Our ground processing system can process the DEM data at a similar rate.
• Automated vegetation discrimination in our ground processing system allows extraction of topographic information for terrain or vegetation surfaces.

PR:    Many of our visitors may not have experience with an actual lidar field survey. Can you briefly explain how DATIS™ is deployed and how it actually operates during a field survey?

RE:    For the majority of field operations, the DATIS™ is typically mounted on small, fixed wing aircraft. This reduces cost as compared to helicopter or larger aircraft platforms. The on-board electronics collects, formats, and stores in-flight laser range, camera, GPS and IMU data for subsequent ground processing. During actual data collection the laser fires at a rate of 4000 pulses per second and is scanned across the aircraft flight path. Since the speed of light is known, the time it takes the laser light to reflect back to the aircraft is measured, which allows the distance to the terrain surface to be calculated. To accurately locate the elevation points, the latitude and longitude of the aircraft are recorded with a high accuracy GPS. The GPS data is differentially corrected using GPS data collected at a high accuracy ground control point. Simultaneously, the aircraft attitude (pitch, roll and yaw) is measured with an IMU and finally the scan mirror angle is recorded. All data is post-processed to create an extremely dense and accurate DEM. First and last laser pulse discrimination is available to support true ground height or feature elevations such as trees, buildings, and other planimetric features. Our process yields up to 220,000 elevation measurements per square mile in hours.

PR:    Earlier you mentioned complementary imaging capabilities; can you give us some more details?

RE:    Certainly. We feel this is a key advantage to DATIS™. If a project has imagery requirements, our digital camera allows us to simultaneously collect high-resolution panchromatic images in addition to the laser data. The camera consists of a 2044 x 2024 pixel array yielding images with 0.3 meter (1.0 foot) pixels in high-resolution mode. The imagery is co-registered and therefore georeferenced to the DEM. Normally these images are first individually orthorectified and then digitally mosaicked together to cover a larger geographic area. Since they are taken in digital form, no film or intermediate process steps are required. Imagery is literally available within minutes after landing the aircraft. Another key advantage of the integrated digital camera system is that a second over flight with a different aircraft and sensor or a standard aerial film camera package is not required to collect image information of the area of interest. This provides tremendous savings to the customer.

PR:    Many of our site visitors are data end-users; I'm sure they'd like to know ..... my this is good egg nog ..... er, like to know more specifics on the DATIS™ system capabilities and mapping products available. Perhaps specifications they could use for determining the suitability of DATIS™ to a particular survey?

RE    There are two basic modes of operation that permit variations in the basic DEM and image specification. The standard resolution mode is well suited to regional planning, utility, route planning, and other large area applications. The high-resolution mode works well for urban planning, site engineering, and other site-specific applications. These two modes of operation have been established to provide a range of data products to meet most of our customer requirements. Additionally we can tailor the system to specific needs as required.

The standard resolution mode provides a nominal swath width of terrain mapping coverage of 1000 meters (3000 feet). Typical elevation data sample postings are approximately every 5 meters (15 feet). The horizontal accuracy of the sample point is typically within 1 meter in Northing and Easting. Vertical accuracy is less than 15 centimeters or 1 sigma. The typical flight altitude for data collection is 1830 meters (5500 feet) AGL (Above Ground Level). The data collection rate is approximately 185 square kilometers per hour and about 14.5 million elevation data samples are measured per hour.

By comparison, the high-resolution mode provides a nominal swath width of terrain mapping coverage of 600 meters (1800 feet). Typical elevation data sample postings are approximately every 3 meters (9 feet). The horizontal accuracy of the sample point is typically within 1 meter in Northing and Easting. Vertical accuracy is less than 15 centimeters or 1 sigma, which is the same as for the wide coverage mode. In this mode, complementary digital images are available with each image having a dimension of 2044 by 2024 pixels and a GSD (Ground Sample Distance) or spatial resolution of approximately 0.3 meters (1 foot). The DEM coverage extends beyond the image coverage to ensure co-registration of the two products. The accuracy of the image is within 2 meters (6 feet) in Northing and Easting. The typical flight altitude for data collection is 2000 meters (6000 feet) AGL. The data collection rate is approximately 110 square kilometers per hour and about 14.5 million elevation data samples are measured and 500 images are collected per hour.

A point to note when comparing to conventional techniques is that, unlike aerial photogrammetry, the entire data collection process requires no normal ground control. Relative positioning GPS is used to reference the data and kinematic GPS post processing is performed to achieve high accuracy. The base station GPS is typically set up over a high accuracy monument or benchmark of known location usually within 80 kilometers (50 miles) of the aircraft location. All data are referenced to WGS-84 datum using relative positioning GPS. The resulting data products are available in other datum as well, based on a customer's need.

PR:    The accuracy of the data produced by airborne laser systems is a hot topic these days, especially in light of the FEMA investigation to qualify airborne laser data for flood plain mapping. How is the accuracy of DATIS™ data measured and verified?

RE:    Well, we have conducted a comprehensive verification of our data accuracy that, unlike other commercial systems, has been verified with several different approaches. For example we completed a metric analysis of the data. In this case an independent analysis was performed on our digital camera and the laser terrain mapping accuracy. Essentially laser mapping accuracy was compared to elevation data extracted from stereo aerial photos with a scale of 1:8000. Unfortunately, the accuracy uncertainty of the photogrammetric-produced DEM was 23 cm, 1 sigma, which exceeded the DATIS™ DEM data error of 15 cm; therefore, no realistic comparison could be made except to state that the laser-based DEM was at least as accurate as the more traditional product. An image accuracy analysis of the same data show DATIS™ imagery can be registered to: 2.0 to 4.0 meters with no ground control, 1.0 to 1.5 meters using block correction with no additional ground control, or 0.7 to 1.0 meters using block correction coupled with additional ground control.

In addition field comparisons have been conducted to determine actual accuracy performance against field measurements. A joint project with Merrick & Company on a power line installation included ground GPS surveys. EagleScan elevation data was better than 4 inches in standard deviation accuracy against GPS survey points. This comparison is shown on our web page. And as part of the Bechtel, Inc., White Sands, New Mexico, rocket test track project, EagleScan elevation data was compared with a series of US Department of Defense, National Imagery and Mapping Agency (NIMA) monuments along the track. The accuracy of EagleScan's data as compared to NIMA data points was better than 5 cm standard deviation.

Each project we fly is inherently a self-calibration test. In order to cover a survey area, multiple flight lines must be flown. Flight line separation is chosen to provide overlapping data to allow for navigation error and aircraft attitude variation. When two flight lines are merged this is equivalent to taking two separate bearings on the same location. These two sightings are statistically independent in that it is 10's of minutes to hours between flight lines. If the system is not aligned accurately, these two independent sightings will not agree resulting in elevation discontinuities at the flight line seams. In addition to the adjacent flight lines, EagleScan's routine operation requires that one flight line be flown in opposite directions as a quality control check. Based on the overlapping data and the quality control flight line any alignment errors can be computed and corrected for in the processing. This self-alignment capability exists at EagleScan as a result of our decision to design and build the hardware and develop the processing software in house. The use of this methodology in effect gives us 4000 ground control points per second.

The limiting factor in the accuracy becomes the quality of the GPS data. This is the reason that our operational times are controlled by the GPS planning models that show the PDOP and we restrict our operation area to within 50 miles of the base reference station.

Finally, whenever we conduct a airborne laser mapping project, part of our standard procedure is to also perform a ground GPS survey. Our surveyors collect a large enough number of ground truth points to provide us statistically valid data to generate a good accuracy assessment of the LIDAR data.

PR:    For people who may be interested in working with EagleScan, can you give us details of your survey operations and deployment region?

RE:    Flight operations, initiated from Jefferson County Airport near Boulder, can support remotely located projects with our current aircraft. Data processing and product generation is conducted at the EagleScan facility located in Boulder. We maintain a full-time staff certified to conduct projects and generate final products for our customers. We have implemented a continuing R/D activity to develop new products and incorporate new technologies into our system and process. EagleScan has a staff of consultants used for various needs including requirement development, government contracting, and hardware and software development. EagleScan has access to consultants with security clearances who have conducted government missions involving data collection using complex electro-optical systems.

The robustness of DATIS™ allows a wider time range for data collection with respect to flight conditions compared with typical aerial photogrammetric systems. Many projects that have been completed to date have been conducted in weather and light conditions that are unsuitable for standard aerial photography. Some missions have been flown in high wind conditions with virtually no impact on accuracy. High quality imagery has been collected in cloudy conditions. The nominal flight altitude ranges from 5500 feet to 6000 feet above ground level (AGL), but this can be lowered to support specific conditions. Since the LIDAR is an active and not a passive sensor, DEM data may be collected at night. To date EagleScan has performed Airborne Laser Mapping™ work throughout the United States during all seasons of the year and may soon begin to take on work overseas as well.