The accurate and most expensive measurements are aimed at changing regionse
The aim of the study was to examine new mapping technology providing much more versatile information about the terrain than current methods. The study creates a vision of how the new data made possible by modern technology can be used and combined with older methods.
In the concept, drawn up by a multinational group of researchers, regional changes – such as new construction and felling of forests – are first observed in open satellite data covering most of the earth. The data concerning the changes is then confirmed using new high-precision methods authenticating the changes’ precise extent and type.
The new technology is much more precise, but also much more expensive, than the methods widely used at present. The research team presents an idea of how at least some of the more accurate measurements can be targeted cost-effectively at significant regions of change. The goal is to have more accurate and up-to-date maps and city models.
“There is a global need to increase the effectiveness of updating methods. Interpretation of change data from the latest remote sensing material can be more thoroughly automated, which opens new opportunities to support map updating with artificial intelligence. In our vision for the future, map updating processes will become faster and more accurate, because mapping experts can focus directly on interpreting changes, rather than looking for them,” says the Finnish Geospatial Research Institute’s Research Manager Eetu Puttonen.
Like going from black-and-white TV to colour
The research is ground-breaking because it is the first to examine the application of a time series of satellite images from the Sentinel-2 satellite, two multispectral airborne laser scans made at different times, and mobile laser scanning to mapping in the same test area. All these remote sensing materials represent the latest technology and provide new opportunities for mapping, which is currently based mainly on aerial imagery and single-channel airborne laser scanning.
Going from single-channel airborne laser scanning – that is, scanning using only one wavelength – to multispectral laser scanning has been compared to going from black-and-white TV to colour. Multispectral laser scanning produces colour intensity images that make it significantly easier to identify different land cover classes and map features compared to black-and-white images.
“Multispectral airborne laser scanning increases the potential for automated mapping techniques. In particular, when multispectral material from two different times is available, it is possible to examine the changes on the ground more closely. In turn, they can be used to find, for example, new roads, buildings and logging sites. In our study, we tested this technology for the first time in the world,” says the Finnish Geospatial Research Institute’s Specialist Research Scientist Leena Matikainen.
What is laser scanning?
In laser scanning, the scanner sends laser pulses which reflect back to it from the environment. The scanning equipment measures the distance to each object from the time it takes for the pulses to return, as well as the intensity of the pulses. A point cloud can then be created from millions of measurements. The point cloud is a detailed, almost photographic, map or 3D model of the environment.
A significant benefit compared to traditional imaging based on photographic technology is the omission of shadows. External lighting conditions do not affect the intensity images, making them more uniform in quality and easier to interpret. Laser scanning can also be done at night.
“The Finnish Geospatial Research Institute of the National Land Survey of Finland is one of the few research organisations worldwide with all the required expertise to conduct similar concept studies within the same organisation. Studies using the latest methods are essential,” Puttonen says.
The research was conducted as part of the Centre of Excellence in Laser Scanning Research funded by the Academy of Finland and a two-year research project under the ICT2023 research programme.
More Information
Research Manager Eetu Puttonen, +358 29 531 4868, firstname.lastaname@nls.fi
See video about the research paper's results
Matikainen, L., Pandžić, M., Li, F., Karila, K., Hyyppä, J., Litkey, P., Kukko, A., Lehtomäki, M., Karjalainen, M., and E. Puttonen, 2019. Toward utilizing multitemporal multispectral airborne laser scanning, Sentinel-2, and mobile laser scanning in map updating. Journal of Applied Remote Sensing, 13(4), 044504.