The Two Best Ways To Capture Data
LiDAR and Photogrammetry have long been critical tools to capture data for surveying and mapping professionals, allowing them to gain a detailed understanding of areas that would otherwise be difficult or impossible to map. This data is critical for land management and urban planning uses, where the ability to reliably measure accurate distance from a piece of topographic data like a contour lines map can mean the difference between the success and failure of an enormous project years in the making.
Traditional surveying methods were time intensive and laborious. Old school tools like the theodolite and transit were used to measure angles and distances between points on the survey site, which would then be plotted on a map by hand. These methods of topographic surveying would often take days just to collect the survey data, and was prone to operator error.
Today, drone surveys have completely revolutionized the way surveying projects are executed. It is now possible to use smaller, lighter and lower-cost LiDAR and Photogrammetry tools. Instead of needing expensive manned aircraft, drones can now be used to capture the survey data. Advanced payloads specifically built for drone surveying are able to perform land surveys and collect data in a fraction of the time and cost of previous iterations, while greatly increasing accuracy.
Both LiDAR and Photogrammetry have proven to be highly successful and efficient methods of drone surveying. In a single automated drone flight, enough aerial data can be collected to generate and directly procure accurate measurements from a digital surface model of the survey site.
While both LiDAR and Photogrammetry offer great advantages, they each have different strengths and weaknesses. In this article, we're going to explain the basics of LiDAR and Photogrammetry, compare their capabilities and discuss when and why they might be the best choice for a given drone survey project.
A Cutting Edge Field
Aerial surveying has remained a field marked by incredible technological growth. From access to satellite imagery to specialized surveying software used to process data, there has been no shortage of new and improved ways to tackle surveying projects.
No tool for aerial imagery has been able to leverage the many advancements like surveying drones. Offering increased camera resolution and providing more strategic methods of capturing aerial images from multiple vantage points, a surveying drone is able to create topographic surveys with high absolute accuracy.
What Is Photogrammetry?
Simply put, photogrammetry is the use of many individual photos being utilized together to measure distances.
Several hundred to thousands of photos are required for efficient photogrammetry. A lot depends on the size of the property and the degree of detail and precision required.
Using specialized software, these images are then processed to create realistic and accurate world models in various formats such as 2D and 3D orthomosaic maps and models. The best photogrammetry drones usually have included software or easy compatibility with existing programs to simplify the matter. These outputs can be used for a variety of purposes, from construction planning to ongoing project management and even in the creation of marketing materials. Photogrammetry is impressive for the immense amount of information able to be generated from a single batch of drone survey data.
What is LiDAR?
Unlike photogrammetry, which relies on traditional photography for data collection, LiDAR (which stands for Light Detection and Ranging) uses lasers for capturing data. The lasers are used to generate a high density point cloud (a set of data points in space) to aid in the creation of digital surface models.
In drone surveying, a LiDAR sensor is mounted on the aircraft and emits short laser pulses as the drone flies over the survey site. These pulses reflect off of the surfaces below and those reflections are then captured by the sensor. Using the time delay between when the pulse was emitted and when the reflection was received, an algorithm is able to calculate the distance to the reflecting surface.
A large number of these distance measurements are then combined to create a dense point cloud which can be used to generate a digital surface model.
As a newer technology, the introduction of LiDAR meant that surveyors would have much more accurate data to work with than was previously possible with photogrammetry.
Which Aerial Data Is Better For Surveying, Photogrammetry Or LiDAR?
Well, it's complicated.
Each of these two drone surveying methods have strengths and weaknesses, and we're going to go into them in detail. But at a top level, photogrammetry is better at capturing fine details, while LiDAR can provide more accurate measurements of larger areas.
How Does LiDAR Work?
As mentioned above, LiDAR sensors use lasers to track the time it takes for light to reflect back from the ground and the intensity of that light.
Despite being around for decades, LiDAR technology has only recently become small enough to fit onto a drone's payload.
Laser measurements are only part of a very complex process. LiDAR uses other high-accuracy technologies such as satellite positioning (GNSS data) and an inertial measurement unit to collect the data needed to create a point cloud accurate enough to properly reflects terrain and its topography (IMU).
LiDAR technology for mapping is known for its precision and accuracy.
In the world of surveying, there are varying types of accuracy, each with different implications as it relates to a project.
Relative accuracy measures a subject's features in relation to another specific subject.
Absolute accuracy, measurements are accurate in relation to the larger world as a whole.
When generating a realistic bare-earth model, LiDAR is the best method to obtain perfect accuracy. This is due to the fact that it delivers the most precise elevation, vegetation, and current condition data.
The best level of vertical precision for your digital terrain model is achieved by combining GNSS data with the usage of LiDAR, which fires hundreds of laser pulses from above.
Topographical issues are caused by more than merely terrain undulations. Photogrammetry and other photo-based surveying methods can be impeded by flora, which prevents finer ground-level readings from being acquired.
LiDAR laser pulses reach the ground below by passing through leafy branches and leaves, improving measurement accuracy.
If the levels of light at your jobsite aren't consistent, LiDAR is a better option. Without the use of external illumination, LiDAR can be utilized to conduct nighttime surveys or low-visibility activities.
LiDAR also helps to collect information on subjects with very small diameters. A good example of this is electrical cables. Due to high-density point collection and a direct measurement approach, LiDAR can precisely map wire catenary.
Working using LiDAR has a number of drawbacks, the most obvious of which being its expense. A complete survey system can potentially cost hundreds of thousands of dollars due to the increased operational complexity (and the requirement for more sophisticated components and data sensors).
This complexity also raises your reliance on an experienced specialist and increases your margin for error. Extracting the drone data you need isn't easy with several sensors and information that isn't immediately accessible without a lot of processing.
LiDAR sensors are typically larger than traditional cameras. Aerial surveys are growing increasingly common, and the need for a larger drone to carry a heavier payload can add to the already expensive cost.
LiDAR's third problem is that it is the best tool for the job in very specialized settings, which is also its greatest strength. For many tasks, standard photogrammetry will suffice. This is a practice that is gaining traction as image processing software improves.
One of the key advantages of using photogrammetry is its accessibility. With drone technology and mapping software, any company with a quality camera drone can create realistic maps and 3D models.
The process of completing a mapping mission and converting the data into useful information is rather straightforward once you've completed camera calibration, basic flight planning, and drawn your ground control points. There are numerous instances where this approach produces concrete results in industries as diverse as building, conservation, mining, and agriculture.
Importantly, various outputs are also available and easily distributed. The simplicity of maps and models with identifiable features and colors makes them a powerful collaboration tool, allowing stakeholders to exchange information without having to spend too much time modifying the information.
Photogrammetry's inexpensive cost is another major part of its appeal. As previously stated, getting started means a modest sum in a camera drone and far less on the data processing software.
Finally, photogrammetry allows us additional flexibility. Based on the task you're working on, you can choose how fast, how high, and how accurate you want your mission to be.
Photogrammetry-based surveying technologies do have a few drawbacks.
The first is that the quality of your drone's camera and the drone itself has a significant impact on the accuracy of your maps and models.
The ground sample distance (GSD) is affected by sensor size, aperture, resolution, and focal length, as well as the altitude at which you are flying. Furthermore, without numerous ground control points or an RTK or PPK-enabled drone, you'll struggle to obtain results with absolute accuracy.
Drone pilots will have to figure out the best flight altitude to achieve the required ground sample distance. To ensure that your software can flawlessly stitch your photographs together, you'll need to set up an overlap on each image.
The weather is the second obstacle in the way of your Photogrammetry goals. Or, to put it another way, the lighting circumstances. The quality of your surveying results can be harmed by factors such as darkness, cloud cover, dust, and more.
It's impossible to measure what you cannot see when using data processing. This means that flights with poor visibility will provide fewer ground points and less accurate maps and models, whether owing to vegetation, shadows, or the time of day.
Photogrammetry vs. LiDAR: Which Is Best?
LiDAR and photogrammetry are two fundamentally different data collection methods.
With LiDAR, you get thousands of data points that build a 3D point cloud that outlines the ground. To make it visually accessible, you'll need to combine color from several different databases.
Photogrammetry produces hundreds or thousands of photos that must be analyzed and stitched together to produce anything useful, such as a 3D point cloud, map, or navigable model.
LiDAR processing in the cloud isn't as common or as accessible as cloud-based photogrammetry tools. That means you'll need an on-site expert who can turn raw data into usable information, as well as the appropriate software.
Which is more accurate: LiDAR or Photogrammetry?
When compared to photogrammetry, LiDAR produces scans with more detail and precision. It's also great for cases when precision is more important than anything else, such as low light or a lot of vegetation, because it can operate well despite environmental constraints like low light or a lot of vegetation.
LiDAR point clouds can have up to 500 points per square meter of granularity and a vertical elevation accuracy of less than three millimeters. A dataset with a high density of data points is more robust, giving you greater flexibility when it comes to processing your findings.
That isn't to claim that photogrammetry is always correct. You can still create highly detailed maps and models if your landscape is relatively simple and devoid of dense vegetation - especially if you use an RTK positioning module.
When Should You Use LiDAR?
If you're mapping difficult terrain with a high proportion of vegetation coverage, LiDAR is a good option. Because of the direct measurements that pierce between leaves, branches, and trees, the data may be used to create accurate topographical point clouds.
The technology is particularly great for precisely measuring items like cables, which are typically too thin to be detected using traditional methods.
If the surveying work at hand necessitates precision above all else, LiDAR should be your tool of choice. Although this is not without its difficulties, which include costs and the knowledge needed to bring the data to life.
LiDAR is ideal for:
- Creating maps of terrain that is difficult to access, intricate, and overgrown.
- Modeling electricity cables, roof edges, and other objects which must be depicted in fine detail.
- Any projects that require a great deal of attention to detail and precision, such as accident reconsctructions.
When Should You Use Photogrammetry?
For those who are new to drone surveying, photogrammetry is a better option due to its price, though having a lower cost than LiDAR isn't the sole advantage.
In fact, Photogrammetry would be a better fit for a lot of applications. This is especially true when you want to collaborate on plans using orthomosaic maps, 3D models, or other affordable project status updates.
Choose Photogrammetry if you want to:
- Create maps and models that are easy to interpret and don't require a lot of post-processing.
- Generate datasets that need to be visually inspected.
Drone Photogrammetry Solutions
The DJI Phantom 4 RTK is an all-in-one, professional surveying solution. The Phantom 4 RTK has an onboard integrated RTK module, which provides real-time, centimeter-level positioning data for better absolute accuracy. The Phantom 4 RTK can also record and store satellite data for use in Post Processed Kinematics (PPK), which may be performed using the DJI Cloud PPK Service. Professional surveyors and beginners alike will appreciate the Phantom 4 RTK's affordability, accuracy, and ease of use.
The P1 is a full-frame photogrammetry payload with adjustable fixed-focus lenses and a full-frame sensor. It's perfect for large-scale photogrammetry flights since it has a global mechanical shutter and software features like Smart Oblique Capture. The P1 uses the M300 RTK to cover 3 km2 in a single flight and produce 3 cm horizontally and 5 cm vertically accurate results without the use of GCPs.
LiDAR solutions from DJI
Similar to the P1, the Zenmuse L1 is a premium LiDAR sensor designed to be utilized with the M300. Integrated with a 3-axis stabilized gimbal, the L1 houses a Livox LiDAR module, a high-accuracy IMU, and a camera with a 1-inch CMOS.
This comprehensive surveying solution offers 3D data as well as RGB imaging capabilities.
The drone can provide high-resolution images of complex structures and precisely reconstruct the images with accuracy of 5 cm vertically and 10 cm horizontally, over an area of up to 2 km2.
LiDAR and Photogrammetry aren't perfectly positioned to be compared apples to apples as two rival data collecting technologies. It's not that one is necessarily better than the other, as we've discussed. The task at hand will ultimately determine the best answer.
Photogrammetry is a good way to capture survey data if the lighting and contrast are good. However, for difficult mapping tasks involving elevation precision, complex structures, or partially obscured terrain, LiDAR is likely the best option.
In any case, the cost and experience of your staff will play a significant part in determining the best method. As DJI's top surveying payloads, the P1 and L1 demonstrate that the technology has become increasingly affordable and accessible.