Bharat Lohani
Research Officer, Environmental Systems Science Centre
The University of Reading Harry Pitt Building, 3 Earley Gate Whiteknights
PO Box 238, Reading RG6 6AL, U.K.
Email: [email protected]
LiDAR is establishing itself as a major player in the topographic data collection industry.
Accurate, timely and reliable topographic data are the backbone of all geoinformation projects. The demand for up-to-date data has increased manifold with the growing operationalisation and implementation of Geographical Information System (GIS). However, the growth in high-resolution GIS applications has not been paralleled by the development of suitable data collection techniques. Until recently, only conventional techniques of topographic data collection (viz. surveying, levelling, photogrammetry, satellite stereogrammetry etc.) were available to the geomatics community. These techniques suffer from one or other limitation in terms of their accuracy, resolution, time, site inaccessibility, weather and sunlight dependence. The recently emerged technique of airborne scanning laser altimetry, generally referred to as LiDAR (Light Detection and Ranging), however, promises to eliminate most of the handicaps of conventional methods and is establishing itself as a major player in the topographic data collection industry.
The concept of LiDAR has been known to surveying professionals since the days of Geodimeter and other electronic distance measuring devices. The idea of using airborne LiDAR to measure terrestrial biomass and relief was first formulated in the late 1960s. The early applications of LiDAR were mostly for measuring bathymetry and forest biomass. These studies, however, recognised several limitations, including the need for an enhanced laser system and most importantly poor geolocation of the sensor position. The development of LiDAR technology stayed dormant due to these serious limitations in the late 1960s and ’70s. These problems were, however, eliminated in 1980s with advancements in laser technology and most significantly in the techniques of aircraft positioning and attitude measurement by Global Positioning System (GPS) and Inertial Navigation System (INS), respectively. Today, state-of-the-art LiDAR instruments are capable of producing 2,000-80,000 pulses per second and geo-referencing these to better than 10 cm in the vertical and 50 cm by 50 cm in the horizontal plane.
  |
| Fig. 1: Classification and removal of the forest cover for generating a DEM of the ground underneath. (Image courtesy of Airborne 1, USA) |
Laser as an Altimeter
LiDAR truly stands for integration of several sensors used in airborne remote sensing industry. A LiDAR instrument typically comprises the laser transmitter/receiver, scanning mechanism, timing electronics, GPS, INS, data storage and processing software. The airborne instrument operates by firing laser pulses towards ground and sensing the return waveform. The round-trip travel time of laser pulse is measured precisely with a high-resolution time-interval counter, which determines the distance of the ground spot from the sensor location. The sensor location is determined by the airborne GPS measurements, which are corrected by simultaneous measurements on a or multiple ground GPS stations using the differential kinematic approach. The INS takes account of the aircraft roll, pitch and heading movements during the operation and applies corrections to the laser vector direction from the aircraft to the ground. Adding the laser range vector to the location of the aircraft gives the coordinates of the ground target. The laser scans the ground using an opto-mechanical device and measures the points along a swath orthogonal to the flight line.
The laser selected for LiDAR mapping operates at or near the 1064 nm or 532 nm wavelengths, which are suitable for topographic and bathymetric mapping, respectively. The LiDAR instrument can be integrated by taking different off-the-shelf components viz. laser scanner, GPS and INS. Some manufacturers also provide complete integrated systems. LiDAR can be operated from a single or twin engine fixed-wing aircraft or a helicopter depending upon the application in hand. A typical operation of LiDAR includes fixing the instrument on an aerial platform, carrying out in situ pre- and post-flight calibration of the instrument, establishing ground control points for differential GPS and flying over the terrain along the pre-defined paths.
Fig. 2: Transmission line mapping. (Image courtesy of TopoSys GmbH, Germany)
Data Processing
Post-processing is carried out to combine the laser range vector with the GPS and INS measurements to yield ground coordinates of each and every ground hit of the laser pulse. The data generated at this stage are ASCII files containing the Easting, Northing and elevation coordinates in WGS-84 system of all the laser target points. LiDAR data may be imported as a Triangulated Irregular Network (TIN) into a GIS or interpolated into a grid for further processing. It is prudent to check the LiDAR data at this stage for any unforeseen errors. Comparison of overlaps between two flight lines and comparison with ground control points, if available, provide information on the quality of the data. LiDAR measurements are further subjected to data processing to generate meaningful information.
Advantages
LiDAR measures the topography with a vertical accuracy of 10-15 cm and horizontal accuracy of 50 cm – 100 cm. The time of data acquisition and processing is very short. On an average, 90 –100 sq. km. area can be measured in one hour. Typical post-processing times are two to three hours for every hour of recorded flight data with additional processing time required for more sophisticated analysis such as target classification. Unlike photogrammetry, LiDAR operates in all seasons, weather and sun-angle conditions. Being an active system, LiDAR can also gather data during night. LiDAR produces a very high density from about 1 point per 20 m to 20 points per sq.m depending upon the instrument type and operating parameters. LiDAR can operate in inaccessible areas as only one or two ground control points are required for the full survey. One of the most important advantages of LiDAR is its digital nature, which enables its direct use in a GIS project and integration with other digital data products. The reflected LiDAR waveform can be sampled for the most significant return or first and last significant returns or at multiple intervals, which provides it an unique capability of measuring the vertical structure of the objects. Besides the height information, LiDAR can also produce an intensity image which has been found, though coarsely, useful in distinguishing different object types. Initially, LiDAR cost was cited as a stumbling block in its wider application. However, with more and more users opting for it and more and more companies offering the service, the cost has come down over the years to the levels of or below the conventional techniques.
Applications
Fig. 3: City model of Bonn, Germany derived from LiDAR data (Image courtesy of TopoSys GmbH, Germany)
Notwithstanding its recently gained maturity, LiDAR has already found several application areas requiring high-resolution topographic data. In addition, as is true with all new technologies, LiDAR has opened up several new application areas.
LiDAR bathymetry makes use of the fact that some of the laser pulse is reflected from the surface of the water and the rest from the bottom. Measuring these two returns can determine the elevation difference between the water surface and the bottom. Another interesting application of LiDAR is in forest mapping. Unlike photogrammetry, LiDAR provides the elevations of both the treetops and the ground beneath them. Even in a dense forest some of the LiDAR pulses travel to the ground, measuring its elevation. The heights of trees can be derived by taking differences of the ground measurements and the laser returns from the treetops. The returns from the treetops or bushes can also be separated leaving only the returns from the ground, thereby producing a DEM of the ground under forest cover (Fig. 1).
The requirement of only a few ground control points has made LiDAR an ideal instrument for inaccessible and featureless areas e.g. glaciers, coastal areas, wetlands, volcanoes etc. Temporal LiDAR surveys of these terrains have been able to locate the changes and validate models predicting the changes. Rapid and accurate mapping for damage assessment after natural disasters, e.g. after hurricanes, earthquakes and landslides, has been made possible with LiDAR technology.
One of the exciting application areas, which has drawn most attention and was hitherto considered difficult conventionally, is the monitoring of transmission lines. A LiDAR is flown along the transmission line to gather very dense measurements of the wires, towers, vegetation intruding into the transmission lines, and topography of the corridor. These measurements allow for the clearing of transmission lines of intrusions, and for carrying out repairs and modifications economically (Fig. 2). Similar corridor mapping for roads, railway tracks, pipelines and waterways, both for the purpose of construction and maintenance, are interesting applications of LiDAR.
Another unique application is to automatically capture the buildings in built-up areas for city modelling purposes (Fig. 3). Urban city models derived from LiDAR data are very valuable for cellular phone companies in planning and monitoring the network. Further, LiDAR DEMs can be used to derive the watershed characteristic of a city and plan the drainage network. LiDAR data is also being used for flood forecasting models, which require up-to-date highly accurate topographic information.
Future Developments
One of the imminent developments awaited in the industry is the pushbroom type operation of laser sensors, which will make the instrument more rugged, light and simpler. Another development in using echo waveform digitisation ranging instead of time-interval counter of the present systems will have the ability to detect the reflectance and to resolve the vertical structure of the surface in much more detail by sampling and storing the whole echo waveform. Parallel to the hardware development, the software tools will become smarter, to automatically extract the useful information from the data and provide better visualisation tools. Furthermore, LiDAR, which essentially provides geometric information, will be integrated at both hardware and software levels with imaging sensors, which are rich in spectral information.
Conclusion
During the last few years, airborne scanning LiDAR has gained increasing acceptance among the geomatics community for collecting high-resolution topographic data. It is an established method now with high technical and economical performance. LiDAR has acted as a complementary data collection technique in some areas, while completely replacing the conventional techniques in others. Further, LiDAR, due to its typical characteristics both in data collection and data type, has opened up several new applications hitherto infeasible with conventional techniques. With further advancements in technology, reduction in the operation cost, and integration with other sensors, LiDAR will become an essential tool for high-resolution geographic information collection.
Acknowledgment
The author is thankful to the TopoSys GmbH, Germany and Airborne 1, USA for permitting use of their data in this article.
