By: Paige Parker, VP of Sales
At my previous job, I was an international business development associate for a LIDAR processing software company. One of my duties was to establish a global customer base. While talking to potential clients, I discovered government budgets were being cut overseas. Several government agencies were switching to a new type of affordable point cloud data also known as SGM (Semi Global Matching).
In 2005, German Scientist, Dr. Heiko Hirschmüller released his research on the process of Semi Global Matching and its results. From then on, this type of data has gained popularity across the globe and was even featured in an ISPRS presentation written by some of the best known companies in the GIS industry.
In layman’s terms, Semi Global Matching is the process of matching pixels from a series of high resolution photos in order to derive a DSM or DEM. This process creates point cloud data, similar to LIDAR. Since the SGM point cloud data is derived from imagery and not from a laser scanner, each data point includes an RGB value, giving a nice visual of an area in real color. Laser data does not usually have RGB values, and in order to get real image color an image overlay usually has to occur or a fusion process.
In most cases, SGM is less dense than LIDAR data. LIDAR provides a better ground model most of the time because of its ability to penetrate vegetation. This gives LIDAR the ability to create more accurate ground models which are essential to processes such as hydrological modeling. SGM can still be a logical alternative for other processes, such as building extraction and working with elevation values, in essence making it a real cost saver. So, depending on the application, SGM is definitely worth considering as an alternative to LiDAR.
Dr. Heiko Hirschmüller stated, “The main power of SGM can be demonstrated on aerial images of urban scenes. A study compared SGM based digital surface models from images of different commercial aerial cameras with each other, with a laser model and ground control points. Very precise DSMs in the ground sampling distance of the images were computed, especially on datasets with sufficient image overlap.”
From a business and sales perspective, SGM offers an affordable alternative solution for people in the industry wanting to work with point cloud data. The cost of SGM is far less than LIDAR for a few reasons. Aerial imagery is typically needed by local county governments. Since they are already investing in imagery they do not have to pay for any extra separate flights to collect point cloud data because the data is directly derived from the images, whereas, LIDAR would be a separate flight. Laser scanners can also be quite expensive adding on to the price of LIDAR along with longer flight times for the acquisition itself, also bumping up cost. So if your budget is tight, considering SGM for your point cloud needs might be worth your time.
If you are interested in getting a cost estimate for your county in regards to SGM data, please contact us at firstname.lastname@example.org.