3D data has been collected and used for over 100 years. What has changed recently is the ability to collect vast quantities of 3D data, to present and interrogate the data in new ways. While BIM has become essential for professionals in construction industry, 3D technology is finding huge utility in defence and internal security, facilities management, urban planning, gaming and entertainment industries. Read on to find out the trends in 3D technology, which is being actively pursued by geospatial industry..
Geospatial information is three dimensional. However, it is often presented in 2D, 3D or even 4D. Instrumentation and software to collect 3D data have changed significantly over the past 10 years. Laser scanning from the air and the ground is the biggest development, but the development of digital cameras and the software to carry out stereo matching has allowed the use of images to compete with laser scanning. Laser scanning and imagery is combined in mobile mapping systems. In addition, the collection of 3D data for database updating, often covering small areas, has been greatly enhanced by the use of GNSS, allowing accurate, rapid collection of data on tablets. Today, 3D information can be visualised dynamically on large computer screens, but users also want information on their tablets and phones, which is not easy.
The introduction of building information models (BIM) in the construction industry is certainly a growing professional application. 3D visualisations have found an important niche in navigation, where the emphasis is on the data being up-to-date rather than being accurate at centimetre scale. A more professional application is mapping of public facilities for inventory and facility management. Today, the use of 3D data is not confined to large scale applications. There is a growing interest in ””””digital globes”””” showing information on both natural and man-made features worldwide. In this case, visualisation is often dynamic and the question of scalability becomes important. Let us explore the trends and look at the direction in which the collection and use of 3D data is moving.
DATA ACQUISITION
Geospatial data has been collected for centuries, initially only in two dimensions. But since the 18th century, principally starting in France and India, elevation data has been collected and shown in various ways. Elevation has been shown as contours, giving an accurate model of relief and through various forms of texturing. The data for these 2D maps has been collected by traditional survey methods: plane tables, tacheometry and later from aerial survey. With the advent of computers, the concept of a digital elevation model (DEM) has been introduced. This allowed computation for earthworks, for example, to be done and even basic visualisation. By the 1960s, DEMs has become commonplace and excellent visualisations were available. The main source of elevation data was then aerial survey, contours were initially drawn manually, or spot heights measured manually, but the development of stereo matching software in the 1980s allowed this to be done automatically. At first, the DEMs generated had errors and gaps but now they are much more reliable, though manual editing is still required. Images come from aerial cameras and satellite sensors. There are several global DEMs available as shown in Table 1:
In the past decade, laser scanning has become an alternative technique of collecting 3D data, both from the air and ground. Images can be taken with the laser data from the air to generate photorealistic orthoimages and mobile mapping systems from the ground combine images with laser scanning. Laser data and DEMs generated from images can be collected separately and combined.
Fully automated generation of very dense 3D point clouds and digital surface models (DSM) from stereo aerial images (nadir and oblique) is rapidly gaining ground. The most efficient method is to use Semi Global Matching (SGM) algorithms. There are a number of software packages which produce accurate point clouds for 3D analysis as well as textured DMSs from imagery collected in flight missions by aeroplanes, helicopters, UAVs, satellite imagery or terrestrial photography collected by mobile mapping vehicles (Table 2).
A new development in mobile mapping systems is Trimble””””s Indoor Mobile Mapping System (TIMMS). It is a manually operated push-cart designed to accurately model interior spaces without accessing GPS. It consists of three core elements: LiDAR and camera systems engineered to work indoors in mobile mode, computers and electronics for completing data acquisition and data processing workflow for producing final 2D/3D maps and models. The models are geolocated, meaning that the real-world position of each area is known. Refer to April 2012 edition of Geospatial World for a more detailed explanation of mobile mapping systems and airborne LiDAR.
PRESENTATION OF 3D DATA
Three dimensional solid models date back to the 18th century and what is claimed to be the oldest large scale mountain relief in the world, created between 1762 and 1786 by F. L. Pfyffer von Wyher can be seen, along with other examples in the exhibition room of the Lucerne Glacier Garden. These models were expensive and only used for special purposes and were essentially for display purposes rather than to provide accurate dimensions. Raw data can be processed by a user with a range of tools. These are improving today but for a long while, there were limitations to the processing possible with laser scanned data. The presentation of 3D data used to be constrained by the media available. Until the 1970s, the medium was paper. The advent of the electronic computer allowed 3D displays of the data; the big breakthrough was the concept of the DTM or DEM, at MIT in Cambridge, Massachusetts in 1956. Originally developed for the terrain, the DEM has now progressed for the representation of anything from surfaces of oil paintings to the surface of the earth and other planets. Not only has the range of scale expanded, but so has the detail shown. 3D visualisations are now widely available for display on a wide range of devices. These range from 3D street views, such as those available from Google and Bing, to view on phones, tablets or large computer screens through to massive scientific models such as used by the Japanese supercomputer centre for environmental modelling.
For simulations to present an accurate picture of the real world, the accuracy of the geodata used is extremely important. Besides, it should be based on the latest available imagery. 3D city and terrain models form the basis of simulations across a plethora of applications and industries.
There are a range of 3D city models available today. ComputaMaps, France, has recently introduced its DxM digital elevation product line, which covers over 275 metropolitan areas worldwide. DEMs serve as the base layer for mapping across a broad range of applications such as urban planning, environmental monitoring, defence and security. LandSIM3D allows the creation of 3D virtual models of entire cities and landscapes by combining geospatial data with 3D or CAD objects in a realtime 3D georeferenced environment. This allows improved decision making related to planning, management and development of sites and territories.
Esri””””s CityEngine is a standalone software that offers professional users in urban planning, architecture, GIS, entertainment and general 3D content production with a conceptual design and modelling solution for creating 3D cities, buildings and streetscapes.
Bentley Map V8i offers the capability to visualise smart 3D models; create thematic maps; capture smart 3D features of buildings, roads and other infrastructure; integrate 3D data from different sources; and thus create realistic renderings and animations.
CityGML is a common information model for the representation, storage and exchange of 3D urban objects and landscape models. It is an OGC standard for presenting real-world features in 3D with different levels of detail. It offers a mechanism for describing 3D objects with respect to their geometry, topology, semantics and appearance. This makes it possible to employ virtual 3D city models for sophisticated analysis tasks in different application domains like simulations, urban data mining, facility management and thematic inquiries.
PLETHORA OF APPLICATIONS
3D digital data is being used in building information modelling (BIM). BIM is an innovative approach to the designing and documentation of building projects. BIM offers the ability to model and manage not just graphics, but also information, which in turn allows the automatic generation of drawings and reports, design analysis, schedule simulation and facilities management. All this gives the building team the ability to make more informed decisions. Besides, BIM supports a distributed team which allows effective sharing of information throughout the lifecycle of the building and eliminates any possibility of data redundancy, data re-entry, data loss, miscommunication and translation errors. An example of the use of BIM is with Crossrail in UK (discussed in Geospatial World in March 2011). In MASDAR City (Geospatial World July 2011), 3D models have been used to estimate temperature rise due to solar heating and structural shapes and to look at the contribution of cities and suburbs to heat dissipation. It appears that tall buildings and urban canyons actually are cooler than open residential areas. The use of 3D models to look at how sound travels is another planning application.
Defence and homeland security: Governments around the world are today faced with monumental challenges tackling the threats to national security. Technology breakthroughs are offering the advantage of low-cost, easy-to-use and realistic visualisation capability for planning, training and decision making. 3D technology is used for diverse purposes in the defence sector including hazardous material safety training, emergency response training, maritime security training, crisis management and war training.
A three-dimensional virtual model of the enemy territory or actual mission place can be created using satellite imagery and emergency procedures can be practiced for various scenarios in advance. Real-time 3D visualisation is a fundamental tool in any situation related to defence and homeland security. Developed by Bohemia Interactive Australia, Virtual Battlespace is a military simulator which uses 3D gaming technology to offer realtime scenario management facilities. The combination of military simulator functionality and modern gaming technology has led to a broad military customer base from around the world including the United States Marine Corps (USMC) and the Australian Defence Force (ADF).
Entertainment industry: Geospatial technology has today become an integral part of the 3D entertainment industry. The use of satellite imagery has gained popularity in the video games. The availability of high-resolution stereo satellite imagery allows users to experience video games in a realistic 3D simulated world.
High-resolution stereo satellite images, together with a terrain elevation model, helps game developers to create a simulation model and visualise the landscape in three dimensions. 3D terrain models are used for map updating and in the creation of 3D city models, which are a prerequisite for generating virtual reality environments.
While these were earlier used to simply visualise the built environment, they are used nowadays as 3D interfaces for refined simulation modelling.
In most cases, models of buildings, vegetation and terrain surface are the major features of interest. LiDAR technology can be used to obtain the DSMs. LiDAR data can further be combined with satellite images to generate DEMs to create a 3D virtual world. In Microsoft””””s aircombat game titled H.A.W.X. 2, satellite imagery from GeoEye-1 has been used to present diverse vistas such as mountains, coastal regions, deserts and some prominent cities like Cape Town, which allows gamers to pilot the fighter jets with amazingly realistic experience.
Nokia””””s Ovi Maps Racing is a location-based racing game that makes use of Ovi Maps to let the users experience the thrill of racing on the tracks through real world cities by utilising maps data by NAVTEQ and the gaming device””””s GPS. The game offers 3D environment of real world cities created using actual site imagery.
3D visualisations from images are used in the film industry with films such as The Matrix and Spiderman making extensive use of photogrammetry in constructing action scenes. Another 3D application is Hawk-eye used in tennis and cricket to determine the path of the ball.
CONCLUSION
The third dimension is an integral part of geospatial information. 3D data has been collected and used for over 100 years. What has changed is the ability to collect vast quantities of 3D data using new technology, and the ability to present and interrogate the data in new ways. There is no doubt that tools such as BIM will become essential for professionals, particularly in the construction industry. Those working on the environment will use 3D models of the terrain, enhanced with environmental information, at all scales from global to local.
3D city models and landscape models are now available, but one would want to know who the users are. Can a single specification satisfy all? How do National Mapping Agencies (NMAs) respond to different requirements? NMAs are handicapped with uncertain or vague possibilities for the utilisation of 3D landscape models. This stops them from deciding the level of detail to adopt. Then there is the problem of accuracy. Is the DTM accurate enough to allow the 3D features to fit into it without misfits? How will data be captured to ensure that there are no gaps and that roads and rivers fit properly to the terrain? Workflows will need to be adjusted to collect 3D data from images, LiDAR or field survey; and staff will need to be trained in the new requirements. All of this suggests that it will be some time before NMAs add 3D landscape models to their portfolio of products.
