Why go for 3D geospatial data from satellite sensors?

Why go for 3D geospatial data from satellite sensors?


Important components of creating a GIS infrastructure that goes beyond two-dimensional space are real 3D vector data, digital terrain models, and information modeling of construction (BIM). The creation of three-dimensional GIS was made possible by integrating GIS methods and technologies, photogrammetry and remote sensing data.

With the advent of digital photogrammetry, stereo images have become the main source of obtaining three-dimensional geospatial data. Photogrammetric processing technologies can be used to create terrain models, obtain 3D-vector data, and create orthophotos. Their use has already become commonplace for engineering, environmental and infrastructure planning, as well as state mapping.

Creation of high-precision geospatial data is quite a complex and expensive process, requiring expenses for collection and processing of source materials, availability of modern software and hardware.

The main advantage of stereo images is the availability of additional geospatial information, which serves as the basis for the production of digital terrain models (DTM) and digital orthophotos.

DTMs are created using software with a built-in autocorrelator and are a key component for the production of digital orthophotos. In addition, DTMs are used to solve a wide range of other tasks.

Digital orthophotos are the main source material for the production of maps, creating and updating vector data, detecting changes.

Another advantage of stereo images is the ability to measure objects on the X, Y and Z coordinates and obtain 3D vector data.


GeoEye-1 satellite (a commercial project of GeoEye Company (USA) was launched in September 2008. It is equipped with two ultra-high resolution sensors: panchromatic with a resolution of 0.41 m and a multispectral resolution of 1.65 m. The images taken from the satellite have a resolution lower than most aerial photographs. Although the main advantage of satellite imagery is a wider coverage of the surface to 210 sq km.

As an example, consider the process of creating different products from a single GeoStereo stereo pair with RPC coefficients obtained from a GeoEye-1 satellite. An important advantage of satellite stereoscopes is that even without reference points, it is possible to create relative stereopairs that will have an accuracy comparable to the original pictures. This is not only convenient for mapping remote areas but also allows you to significantly reduce costs.

To obtain geospatial data from stereo images, digital photogrammetry systems are used, for example, Leica Photogrammetry Suite. The process of obtaining data includes the selection of project parameters, the input of coordinates of control points, the automatic arrangement of tie points, the phototriangulation, the creation of a terrain model, and the orthorectification of the images. Note that you can extract 3D data immediately after performing the phototriangulation.

The project configuration includes the selection of a geometric model (GeoEye RPC) followed by the addition of snapshots to the LPS Project Manager. At this stage, pyramidal images of the images can be created.

Next, you enter the ground control points (GCP – ground control point) in stereo or mono modes using the Point Measurement Tool. This module allows you to associate the file/pixel coordinates of the snapshot with the real coordinates of the GCP. As a rule, this stage is quite laborious. LPS has an “automatic XY drive” function that moves visible snippets of snapshots to an approximate location of the GCP. After the GCP is defined, automatic tie point measurement can be started. In the absence of ground control points, tie points are needed to create a relative stereopair. The next step will be phototriangulation, the results of which serve as the basis for creating digital terrain models and 3D vector layers.

Then, using the LPS Automatic Terrain Extraction module, a digital terrain model is created, which later not only serves as a basis for orthorectification but can also be used to solve other problems.

The result of the subsequent editing of the terrain model with the LPS Terrain Editor module will be a digital terrain model (DEM). Editing the surface is necessary to eliminate errors that can cause distortions in orthophoto planes.

Further, using DTM, ortho correction of pano-chromatic image GeoEye-1 with a resolution of 0.5 m is carried out.

The following modules are used to select 3D objects: PRO600 Fundamentals, designed to work in Bentley Microstation software and Stereo Analyst software for ERDAS IMAGINE software. After extraction, 3D objects can be exported to XML format for later posting to Google.

The obtained geospatial data have a different application. Orthorectified images can be used as a basis for creating GIS, and also used for modeling and visualization of territories together with additional 3D-data.

Also Read

NASA satellite imagery shows the impact of hurricane Florence