In the aerial survey industry, we're throwing phrases like orthorectification, aero-triangulation, and auto-correlation around us, expecting or assuming that the rest of the world knows what we're talking about.
We all have our specific words and phrases in most industries, I guess, so perhaps I should start from the beginning.
When I talk about an aerial survey, I'm talking mainly about the aerial photograph of a piloted fixed-wing aircraft. In recent years, this has become Commonly thanks to Bing Maps and Google Earth, and so on, despite many users assuming that it is all satellite data. Imagery. It's also overshadowed by better sensors such as hyperspectral, LiDAR, infrared, and Unmanned aerial vehicle.
Aerial surveyors fly across the country, taking hundreds of thousands of aerial photographs with highly expensive and specialized scanning cameras. These photos are considered the raw materials for a host of other derived products that are used for applications as diverse as mapping, urban planning, flood modeling, and wind farm modeling.
Each photo overlaps with the previous photo in a good way, to allow us to view it in 3D mode (individual photo is over 1 GB in size as the camera is 262 megapixels). The overlap is usually between 60% and 80%).
Using mathematical calculations based on the camera's information on the aircraft's position and tilt, we triangulate the photos. Shortly, this process turns each photo into a 'map' by setting it in its correct geographical location. That's named an orthophoto and is the kind of thing you see on Google Earth or Bing Maps; it's not a big satellite image, consisting of thousands of individual aerial photos stitched together.
The ortho-rectification process has critical issues, such as removing the distortion caused by the earth's surface or terrain. Still, if you think the top of the mountain is closer to the camera than the sea is, this distorts the scale. The center of the image is right below the camera, while the photo edges are seen at an extremely slight angle that causes distortion again. like showed in figure 1
We create a 'digital terrain model' (DTM) of overlapping images to remedy this and use it to remove the distortion caused by the ground shape (hills and valleys). This process effectively removes distortions caused by hills and valleys but doesn't do the same for buildings; As for the edges of the image, they are slightly viewed from the side. figure 2
This is commonly referred to as 'building lean' and has historically been very costly and difficult to remove in aerial photos, especially on a wide scale. The lean building's primary issue is that the building top is relocated so that the ground and detail behind the building can be obscure. This is getting more defective as the building gets taller. Another issue is that the buildings' tops are not in their correct geographical location compared to the 2D map. This often limits the use of ordinary orthophotos in urban areas.
With advancements in camera and processing technology, everything has changed. Newer digital survey cameras, such as the UltraCam Eagle, are so clear and precise with better positioning technology and optics that we can now produce detailed height models that include buildings, trees, and hills, and valleys. We are currently creating Digital Surface Models (DSM) as standard instead of creating Digital Terrain Models (DTM). Improved camera technology and the fact that the cameras are digital make it possible to raise overlaps from 60 % to 80% or even 90% without extra flying costs.
In "true ortho," the distortion correction level is very high because of increasing overlap; thus, we can photograph each point on the ground multiple times and improve the correlation between pixels. This is because of that DSMs are so good and computers so powerful that they can correct the displacement of individual buildings and the hills and valleys as in ordinary orthophoto.
True Ortho is not considered a new concept. Until now, it has been a manual and highly specialized process. Therefore, for most surveys, it was prohibitively costly and remained within the expert realm. This new technology releases True Ortho's potential to every user.
Removing the obscurations (obscuring buildings and other non-natural features) reveals more of the photo so that every single object (including the roof detail) is in the correct position. We can now use orthophotos to create and update topographic maps and urban plans more precisely. Utilities, Streets, street furniture, pavements, and other similar features are revealed, which significantly improves mapping, inventory, and asset management. The image itself is more aesthetically pleasing and can be considered a bona fide 2D map.
Besides, the DSM, which is a by-product, is a beneficial data set. Its applications are so many and diverse, including the conversion to a 3D mesh and the creation of 3D city models, which were also widely used in urban modeling, planning, sightline, and flood assessment. The earth's surface, both natural and non-natural, has never been so easy to visualize in 3D across such expansive geographies.
In short, the survey and production of orthophoto have improved, with no increase in acquisition costs. The resulting true ortho maps are now more commonly used as the norm as they can now coincide with comparable vector mapping like the Ordnance Survey MasterMap for the first time. At the same time, the DSM produces new and novel applications in the sectors, from insurance to renewables and security to tourism.
Comparison between Orthophoto and True Orthophoto
The "regular" orthophoto tumbling of images is a prevalent phenomenon that appears with, e.g., multi-story buildings, especially in edge areas. When these images are overlayed with cadastral maps, There are visible differences between the maps. Precisely the contrary, In True Orthophotos, not only near-ground objects are accurately depicted but also multi-story buildings and engineering structures such as bridges and tunnels.
Advantages of TRUE ORTHO
• No tumbling of images
• No distortions in engineering structures