Photogrammetry has many applications across various fields. These include land surveying, real estate, construction, engineering, forensics, and architecture. Read on to learn more about photogrammetry and the answers to these questions:
- What is photogrammetry?
- How does photogrammetry work?
- What is this measurement method used for?
- What are the types of photogrammetry and their branches?
- What are the applications of photogrammetry?
This article also touches on the software and hardware needed to make the most of this technology.
What Is Photogrammetry? A Brief Overview
As defined, photogrammetry is the process of getting reliable information about an object, structure, or location using photographs.
The word “photogrammetry” has its roots in the Greek words “photo,” or light, “gamma,” which means “drawing” or “writing,” and “metron,” which means “to measure.” Photogrammetry has been around for centuries, but recent innovations in camera technology and aerial photography have improved the practice and made it even more ubiquitous.
Photogrammetry, in the simplest term, is the process of photography, but in reverse. While taking a photo renders a 3D object into a 2D image, photogrammetry turns a series of 2D images into a 3D model.
Generally, photogrammetry software is used to extract information, such as the elevation, curve, distance, and area, from photographs. Then, the software analyzes overlapping images taken from several angles and locations, gathering, analyzing, and converting data into a 3D digital model. In this way, photogrammetry makes it possible to scan very large objects and landscapes that would otherwise be impossible to capture.
How Does Photogrammetry Work?
The input in 3D photogrammetry is photographs. The output is typically measurements used to create 3D models, maps, or drawings of an object, structure, or area. For example, Google Earth uses photogrammetry to create 3D images of locations.
Take, for example, a picture of an apple on the ground with a ruler beside it, photographed from above. Even with the ruler in the photograph, it is impossible to determine just how big or wide the apple is because the 2D picture does not show whether the ground is curved.
To ensure accurate measurements, we should factor in the curvature of the ground. A traditional photograph will not allow us to do this, but 3D photogrammetry can.
Triangulation: The Primary Principle of Photogrammetry
Triangulation is a technique for determining the relative position of two or more points. For example, photogrammetry takes several overlapping pictures of the apple from different angles and positions.
We can identify points in the image and, by comparing the points in the pictures and taking into account variables such as the position of the camera and its orientation for each photograph, we can determine where these points are situated in 3D space.
When we identify one point in at least two pictures taken from different locations, we can determine “lines of sight” from the cameras in the locations to the point. Then, the lines are mathematically intersected. In turn, this provides the coordinates of the point. Finally, we can create a 3D model of the object with enough points using photogrammetry software.
Scale and Orientation in 3D photogrammetry
Photogrammetric models have proportion, but the scale is also necessary. In order to scale a 3D model, it is crucial to have at least one known distance. This is similar to how people use a familiar object, like a coin, as a reference to better estimate the size of another object.
The equivalent of a coin in 3D photogrammetry is a scale bar. Scale bars have targets on them that are separated by a known distance. This means they can be used to scale an image. Furthermore, because they are coded, they can be read by photogrammetry software and used to determine the orientation of the 3D model.
To ensure accurate and consistent results, scale bars are made from materials that do not change too much in size or thickness, even with temperature variations. Some examples are aluminum, carbon fiber, and steel.
Types of Photogrammetry
Terrestrial Photogrammetry
Terrestrial photogrammetry, also known as close-range photogrammetry, is most commonly used with tripod-mounted or handheld cameras. This type of photogrammetry is most commonly used in non-topographic, meaning it is used for localized measurements, 3D model creation, and point cloud output rather than topographic products like terrain models or maps.
In this process, a camera is used to take photographs to capture the part definition through fiducial targets. Photos are taken from multiple angles.
The camera used may be an ordinary camera. More often, however, it is a specialized wide-angle camera modified to detect reflective-based targets.
Aerial Photogrammetry
One of the most common types of photogrammetry involves taking aerial photographs. A camera is mounted on the underside of an airplane or, if photos need to be taken from a low flight path, on a small, remote-controlled aircraft such as a drone.
The flight path typically goes back and forth over one area. As the aircraft flies along the flight path, the camera takes overlapping photos from several angles, which are then processed in specialized computer software called a stereo plotter.
A stereo plotter allows the user, such as a cartographer, to see and compare two photographs simultaneously. This allows them to determine the complete picture of the landscape or object, such as its height and surroundings. 3D photogrammetry software is then used to process this information and create 3D models.
Aerial photogrammetry is often used for topographical surveys, land-use planning, surveillance, construction progress monitoring, and archeological mapping sites. In addition, it helps capture hard-to-access or dangerous areas where surveying by land is difficult or even impossible.
Space Photogrammetry
This type of 3D photogrammetry involves taking photos using cameras positioned on satellites, other planets, the moon, or on earth. Space photogrammetry, used in space exploration in the 1960s, continues to be valuable today. For example, NASA uses space photogrammetry in shuttle damage assessment.
Capturing Data for Photogrammetry
For photogrammetry to work, detailed photos with sufficient information about the object are necessary. The number of images needed depends on the size and complexity of the inspected geometry.
For instance, the concept behind photogrammetry for engineering is metric photogrammetry. In this branch, the focus is precision.
It involves calculating and gathering precise measurements from photographs by determining the relative x,y, and z positions of placed fiducials within the inspected areas. Images used in metric photogrammetry are typically taken by designed cameras modified to capture light return from the fiducial targeting. These cameras are referred to as metric cameras.
Metrology Engineers will place known targets called “codes” that will assist in recognizing points in the diverse picture positions. These codes represent a pattern of points known to the photogrammetry software; these codes allow for quick pre-alignment as they act as commons within the images.
Once the pre-alignment is completed, a bundle adjustment is performed on the observed fiducials to optimize the picture station positions for optimal accuracy. The final output is a point cloud model resembling the object’s skeleton. In most cases, this point cloud is spare as it only reflects fiducials of interest in the field of view.
Photogrammetry Hardware and Software
The results of photogrammetry depend primarily on the quality of the images used. Simply put, you need clear, detailed photographs to ensure reliable measurements and accurate models.
To achieve this, you need various types of equipment. These include:
- Specialized cameras (metric cameras)
- Passive and Active Targeting
- Scale bars
Since specialized gear for photogrammetry can be expensive, a cost-effective option is to rent it from a trusted provider instead.
Meanwhile, you will also need to use photogrammetry software to process and analyze images and output the required dataset. There are many available online, ranging from free, beginner-friendly applications to more complex tools that cost thousands of dollars.
Which types of software you need depend on your project goals and budget. However, keep in mind that free software typically has limited functionality and fewer features compared to paid applications.
Reliable Integrated Metrology Solutions
Backed by more than three decades of experience in metrology and robotics, ATT is a trusted provider of integrated metrology solutions, including photogrammetry systems.
To learn more about ATT’s products and services or to request a quote, feel free to get in touch today.