The Turkish city of Istanbul is developing a 3D city model mainly aimed at urban planning. The data sources used so far include airborne Lidar, aerial images and 2D maps containing footprints of buildings. Everybody engaged in creating 3D models of large cities faces many issues, challenges and limitations, including excessive data storage requirements, the need for manual editing, incompleteness and other data quality problems. In this article, the authors share their experiences on creating models of the city of Istanbul at the level of detail (LOD) 2 and 3.
The core data for creating the 3D city model of Istanbul was collected by a helicopter flying at a height of 500m and a speed of 80 knots (150km/h) during surveys carried out in 2012 and 2014. The helicopter was equipped with a Q680i Lidar system from RIEGL (Austria), a DigiCam 60MP camera, an AeroControl GNSS/IMU navigation system and an IGI CCNS-5 flight management system. The Lidar point cloud was captured with an average point density of 16 points/m2. The images were recorded with a ground sampling distance (GSD) of 5cm and with 60% along-track and 30% across-track overlap. To ensure high geometric accuracy, eight GNSS base stations were used. The recording of the whole city covering 5,400km2 required thorough flight planning as the flying height and overlap determine a major part of the data quality. Added to this, data accuracy is directly affected by how good the boresights of IMU, GNSS and camera are calibrated and remain stable during the surveys.
The Lidar point clouds were geometrically corrected using RIEGL RiPROCESS and RiANALYZE. The Lidar point cloud was stored in 17,000 LAS files, each covering an area of 500m by 700m. Next these LAS files were used for generating a digital surface model (DSM), a digital elevation model (DEM) and a DSM in which the heights of buildings and other objects refer to the ground surface instead of a local or national reference system. Such a DSM is called a normalized DSM (nDSM). The GSD of the three digital terrain models was 25cm. Combining the DSM with the simultaneously recorded images enabled the creation of orthoimages. The LAS files were also used for classifying the points on the ground, on buildings and in low, medium and high vegetation using MicroStation V8i Connect, TerraSolid and TerraScan software.
The automatic classification of points was done with 90% accuracy. Automatic classification faces severe limitations when adjacent objects clutter. For example, the wrong class was often assigned to buildings close to high trees (Figure 1). Extensive visual checks and manual editing was required to improve the quality of the classification result. Next, the classified points were combined with building footprints extracted from the 1:1,000 base map and with the orthoimages using TerraModeller software to automatically generate around 1.5 million building cubes. Cubes are a 3D representation of level of detail (LOD) 2 (see side bar). Next, the building blocks were augmented by automatically adding roofs using TerraModeller. Mosques, churches, bridges and other complex structures had to be manually mapped using ZMAP software, however. The base map was created from aerial images. Since roof outlines are mapped rather than the actual building footprints, it is not always possible to separate roofs from high trees automatically, thus again requiring extensive manual editing.
The files generated from processing Lidar point clouds, aerial images and 2D base maps were converted to CityGML using an FME Workbench. Next, topologically and semantically correct LOD 2 3D models of buildings were created – in total 1.5 million – with the help of CityGRID software (Figure 2). Various classifications and automatic and manual corrections were made until the 3D model contained the desired details. Based on architectural 3D CAD files (Figure 3), 3,800 landmarks such as mosques were modelled with greater geometric details of facades and roofs (LOD 3) than other buildings. Since no georeferenced photos taken at street level were available, no photo texture was draped over any of the buildings, including the landmarks (Figure 4). The non-textured LOD 3 models were based on the CityGRID XML format to facilitate the topologically correct outlining of roofs, facades, footprints and details such as balconies, dormers and chimneys. In a next step, the created files must be converted to a full CityGML structure.
The creation of the 3D city model of Istanbul is still work in progress. Presently, the main data sources consist of an airborne Lidar point cloud, simultaneously recorded aerial images and the building footprints from the 2D base map of Istanbul. Ground-based data collection has been scheduled to increase the level of detail, with respect to both the geometry and the image texture.>> The preferred technology is laser scanning and 360⁰ panoramic imaging simultaneously captured from a moving car. Many streets in downtown Istanbul are small and narrow and thus inaccessible for cars. It is planned to capture these parts of the city with backpack mobile mapping systems. For the whole of Istanbul, the ground-based data will cover 32,000 kilometres of roads and streets resulting in 2.73 petabytes of panoramic image data. The 3D city model is not yet connected to a database containing semantic building information, but this is part of the future development work.
Biljecki, F. (2017). Level of Details in 3D City Models. PhD thesis, TU Delft, The Netherlands, 353 p.
Buyuksalih, G. (2015). Largest 3D city model ever – case study: Istanbul, Turkey. User presentation at RIEGL Lidar 2015, Hongkong-Guangzhou, China.
Kolbe, T. (2015). CityGML goes to Broadway. Photogrammetric Week 2015, Stuttgart, Germany.
Prandi, F., Devigili, F., Soave, M., Di Staso, U., and De Amicis, E. (2015). 3D Web visualization of huge CityGML models. ISPRS Archives Vol. XL-3W3, pp. 601-605
All the efforts and help on data collection and processing received from the BIMTAS colleagues – Mr Serdar Bayburt and Dr Ismail Buyuksalih especially – is greatly acknowledged. Thanks also go to Mr Hanis Rashidan for the design of Figure 4.
Figure 5: Proposed system architecture of the 3D spatial database and sub-systems.
Last updated: 10/01/2019
http://www.GIM-INTERNATIONAL.com Jueves 11 de Enero del 2019
Providing water for drinking, irrigation and power, glaciers in the world’s highest mountains are a lifeline for more than a billion people. As climate change takes a grip and glaciers lose mass, one might think that, lubricated by more meltwater, they flow more quickly. However, satellite images from over the last 30 years show that it isn’t as simple as that. A paper published recently in 'Nature Geoscience' describes how a multitude of satellite images reveal that there has actually been a slowdown in the rate at which glaciers slide down the high mountains of Asia.
High-mountain Asia stretches from the Tien Shan and Hindu Kush in the northwest, to the eastern Himalayas in the southeast. The area is also part of what is known as ‘the third pole’ because these high-altitude ice fields contain the largest reserve of freshwater outside the polar regions. The source of the 10 major river systems, the third pole provides freshwater for over 1.3 billion people in Asia – nearly 20% of the world’s population.
For more than a decade, satellite data have documented that Asia’s high-mountain glaciers are thinning and losing mass owing to melting. “However, it has not been entirely clear what this loss of ice means for their rate of flow,” said Lead author Amaury Dehecq from NASA’s Jet Propulsion Laboratory (JPL) and from the University of Edinburgh.
Scientists need to understand what regulates glacial flow speed to predict how meltwater will affect the region’s supply of freshwater in the future, and how meltwater adds to sea-level rise.
The study, which was started under ESA’s Dragon programme, uses images from the US Landsat satellites, which are ESA Third Party Missions. Dragon is a joint undertaking between ESA and the National Remote Sensing Centre of China that promotes the use of ESA, Third Party Mission, Copernicus Sentinel and Chinese satellite data for science and applications.
Third Party Missions are not ESA satellite missions, but under an agreement, the data from these missions is also processed and archived by ESA's multi-mission ground systems. The US Geological Survey and NASA Landsat missions fall under this agreement.
Two million pairs of Landsat satellite images
Dr Dehecq and his colleagues analysed almost two million pairs of Landsat satellite images gathered between 1985 and 2017 and used automatic feature tracking to measure the distance that distinctive features on the glaciers, such as crevasses or patches of dirt, had travelled between earlier and later images. Alex Gardner, also from JPL, added: “We did this millions of times to see changes in velocity in the order of one metre a year.”
They found that ice-thickness outweighs any other factor in regulating flow – the thinner the glacier the slower it flows. This challenges the more intuitive theory of glaciers flow quickens because of the lubricating effect of meltwater at their base. One of the reasons for this slowdown is gravity.
Gravitational pull is related to mass, so as a glacier loses mass, the pull weakens causing it to flow slower. Likewise, in the few locations where glaciers have been stable or where they are thickening rather than thinning, flow speeds have been increasing slightly.
Monitoring remote regions from space
Noel Gourmelen from the University of Edinburgh said: “What's surprising about this study is that the relationship between thinning and flow speed is so consistent. These findings should help us better understand how glaciers behaved in the past and better project their contribution to water availability and sea level as they respond to climate change."
“Being able to monitor these remote regions from space over long periods of time is extremely important to understand what is happening. We also now have Europe’s Copernicus Sentinel missions, which also play a crucial role in this kind of monitoring.” With such a large portion of the world’s population dependent on water from these cold heights, changes in the size and flow of these glaciers can bring serious consequences for society.
It is clearly important to continue monitoring this fragile region as recently highlighted by the World Meteorological Organization (WMO) in the National Science Review: Scaling the peaks for social benefits. In recognition of the need to address ease of access to reliable, policy-relevant information on water resources, by integrating emerging knowledge on the accelerated changes in high-mountain cryosphere, the WMO High Mountain Summit will take place in October 2019.
http://www.GIM-INTERNATIONAL.com Martes 18 de Diciembre del 2018
Septentrio, a leader in high-precision GNSS technology, is teaming up with Point One Navigation, a provider of precise location as a service, for autonomous vehicle demonstrations during CES 2019 in Las Vegas, USA. Invitees will be able to ride in a fully autonomous demonstration vehicle that incorporates technology from both companies and to meet directly with technical experts.
Point One Navigation will showcase its proof-of-concept autonomous vehicle equipped with the FusionEngine vehicle localisation software. Demonstrations will utilise corrections from Point One’s Polaris Cloud, an innovative new correction network that enables high-precision GPS and computer vision-based localisation, while allowing the customer to choose the performance and price point that best fits their application. For users operating in open sky scenarios, a Septentrio RTK receiver can be used directly with Polaris Cloud to provide centimeter-level accuracy in seconds. In more challenging urban environments, Point One’s FusionEngine software further integrates cameras and additional sensors to achieve the desired level of precision, even in the complete absence of satellite signals. With coast-to-coast coverage, Polaris Cloud provides a widely available solution for precision localisation today.
High-precision GNSS and computer vision
Point One Navigation has chosen to work with Septentrio to power its solutions for both the correction network and our FusionEngine reference design because of the excellent quality, robustness and jamming resistance of their GNSS receiver technology, said Aaron Nathan, CEO & co-founder. His company has worked with Septentrio before in the autonomous space, and their GNSS technology combined with excellent application support has been integral to Point One’s success, he added.
Neil Vancans, vice-president of global sales for Septentrio stated that Point One Navigation’s thorough understanding of high-precision GNSS integrated with computer vision in a novel way offers a unique approach to solving the challenges in building safe solutions for increasingly autonomous vehicles.
Integrating camera, inertial, and GNSS data
The robust, globally accurate position and orientation obtained by Point One’s FusionEngine software is a result of highly advanced algorithms that integrate camera, inertial, and GNSS data at over 200 times per second. FusionEngine has the accuracy and the resilience to inclement weather demanded by Level 2 applications such as highway lane keeping and V2X, while offering the robustness necessary for mission-critical Level 4 and Level 5 robotaxi and full autonomy applications. What’s more, FusionEngine powered by the Polaris Cloud provides convergence to centimeter accuracy in seconds, not minutes. The platform also completely eliminates the combined location/map provider lock-in typical of other solutions based on HD mapping technology, enabling vehicle OEMs to choose the best map provider for a given region. Point One’s FusionEngine software and Polaris Cloud is a highly innovative combination for safe, cost-effective autonomous driving.
Point One’s solution is powered by Septentrio’s GNSS receivers. For accurate positioning of autonomous vehicles, Septentrio utilises at least two frequencies broadcast by each GNSS constellation (BeiDou, Galileo, GLONASS, GPS, QZSS). With the increased number of GNSS signals obtained from multiple constellations, ADAS-equipped and autonomous vehicles achieve increased accuracy, along with greater reliability, integrity, and availability of GNSS signals.
GNSS module Septentrio recently announced its new mosaic compact multi-constellation GNSS Receiver SiP (system-in-package) module, available in 2019. The Septentrio mosaic, a multi-band, multi-constellation receiver in a low-power surface-mount module with a wide array of interfaces, is designed for mass market applications like robotics and autonomous systems. The mosaic module integrates the latest GNSS and RF ASIC technology, as well as the robust positioning engine from Septentrio.
All Septentrio GNSS receivers and modules feature AIM+ technology, an advanced on-board interference mitigation commercially available. Septentrio GNSS receivers can suppress the widest variety of interferers, from simple continuous narrowband signals to the most complex wideband and pulsed jammers.
Septentrio and Point One Navigation will conduct live autonomous vehicle demonstrations for industry representatives attending the International CES in Las Vegas.
http://www.GIM-INTERNATIONAL.com Martes 18 de Diciembre del 2018
Ultra-high-resolution aerial photography and 3D landscape models from Bluesky are helping Outfield Technologies to develop innovative image recognition techniques to improve agricultural land management and crop production. In addition to the pioneering research project, the Bluesky data is also providing estate managers with up-to-date and accurate mapping for subsidy applications, planning proposals and corroboration of existing data and reports.
Supported by the Eastern Agri-Tech Growth Initiative, with funding from the UK Government Local Growth Fund, Outfield commissioned Bluesky to capture 7,000 hectares of 5cm-resolution photography. Covering farmed land, environmental focus areas and woodland on a large estate in East Anglia, the detailed record of land usage and crop types, for example, is helping with the development of machine learning algorithms to accurately identify features across different datasets and landscapes.
Combining neural networks with geospatial data
Outfield Technologies used a host of analysis software including off the shelf, open source and proprietary packages on its platform to derive useful data from drone images and more traditional aerial survey data, commented Jim McDougall, commercial director of the company. By combining neural networks for image recognition with the Bluesky data they are developing an automated classification process to help improve decision making for a range of applications.
Bluesky also supplied Outfield with a photogrammetrically derived 3D model of the estate, a Digital Surface Model or DEM, which includes the earth’s surface or terrain and landscape features including buildings and vegetation.
This is a funded research project with a high profile, continued McDougall, and there were also great expectations from the estate for the deliverables derived from the Bluesky data. With little margin for error, it was important that the data collection partner fulfilled not only in terms of competitive pricing but also in quality of data and responsiveness of service.
Useful information derived from aerial imagery
This land assessment tool is being trialled to add to Outfield’s existing horticulture management systems. Outfield is a technology company delivering plant level image services for field vegetable growers and orchard producers. By providing these growers with useful information derived from aerial imagery, Outfield helps to improve yield forecasting and sales, maximise outputs and identify problems within a crop.
Outfield’s surveying services can offer significant savings compared to traditional foot surveys and the data can be integrated with farm management systems for recording hedgerows and field boundaries, tackling field margin encroachment, precise measurement of net farmed and cropped areas and assessment of woodland areas.
http://www.GIM-INTERNATIONAL.com Viernes 07 de Diciembre del 2018