Drone Surveying Glossary

The process of collecting data from an airborne platform, typically a drone or aircraft, to create maps, 3D models, or inspect infrastructure. Drone-based aerial surveys are faster, cheaper, and safer than traditional ground surveys for large areas.

The process of calculating the position and orientation of each aerial photo in a dataset relative to ground control points. Aerial triangulation (AT) is a critical step in photogrammetry that ties the 3D model to real-world coordinates.

A widely used photogrammetry software package that processes aerial images into orthomosaics, 3D models, and point clouds. Other popular options include Pix4D, DroneDeploy, and WebODM.

The height of the drone above the ground surface directly below it, not above sea level. AGL is the standard reference for flight planning because it determines image resolution (GSD) and legal compliance.

A digital elevation model with all vegetation, buildings, and other surface objects removed, showing only the natural ground surface. Bare-earth models are typically produced from LiDAR data by classifying and removing non-ground points.

An optimization process in photogrammetry that simultaneously refines camera parameters, 3D point coordinates, and image positions to produce a consistent 3D reconstruction from multiple overlapping images.

Ground control points that are not used in the processing of a photogrammetric model but are reserved to independently verify the accuracy of the final output. Checkpoints provide an unbiased measure of model quality.

An FAA system that allows drone operators to request and receive near-real-time airspace authorizations for controlled airspace. LAANC is essential for legal flights near airports and in Class B, C, D, and E surface areas.

The use of repeated drone surveys over time to track construction progress. Orthomosaic maps and 3D models captured at regular intervals let project managers compare actual progress against plans, measure earthwork volumes, and document site conditions.

A physical marker on the ground with known coordinates, used to georeference aerial data. Ground control points (GCPs) are typically placed before a flight and measured with RTK GPS rovers to ensure survey-grade accuracy.

A 3D representation of terrain elevations as a raster grid. A DEM can represent either the bare earth (DTM) or the surface including objects (DSM). DEMs are used for contour generation, watershed analysis, and site planning.

A digital elevation model that includes the tops of all surface features such as buildings, trees, and infrastructure. A DSM represents what the drone sees, while a DTM strips these features away to show only the ground.

A digital elevation model representing only the bare ground surface, with vegetation and structures removed. DTMs are essential for engineering design, flood modeling, and cut/fill calculations.

A cloud-based drone mapping platform for flight planning, photogrammetric processing, and data sharing. DroneDeploy is popular in construction and agriculture for its ease of use and collaboration features.

A digital surface model (DSM) includes the tops of all objects on the terrain, while a digital terrain model (DTM) shows only the bare ground. The difference matters for engineering: a DSM would show a building on top of the ground, while a DTM shows the ground elevation beneath it.

A popular enterprise drone for mapping and inspection, featuring a mechanical shutter, RTK compatibility, and a 4/3 CMOS sensor. The M3E is commonly used for topographic surveys, inspections, and photogrammetry missions.

The FAA regulation governing commercial drone operations in the United States. Part 107 requires operators to pass an aeronautical knowledge test, register their drone, and follow operational rules including visual line-of-sight, daylight operations, and maximum altitude of 400 feet AGL.

The waiver provision under Part 107 that allows commercial drone operators to fly in controlled airspace near airports. Operators must obtain authorization through LAANC or the DroneZone portal before flying in Class B, C, D, or E surface airspace.

A pre-programmed mission that tells the drone exactly where to fly, what altitude to maintain, how much overlap between photos, and what camera settings to use. Good flight planning is the foundation of accurate aerial surveys.

The percentage of image overlap between consecutive photos along the flight direction. Standard photogrammetry requires 75-85% forward overlap so that every point on the ground appears in at least three images for accurate 3D reconstruction.

The process of assigning real-world coordinates to aerial data so it aligns with known geographic reference systems. Georeferencing uses ground control points or direct georeferencing (RTK/PPK) to tie the model to a coordinate system like UTM or State Plane.

A visible marker placed on the ground with precisely surveyed coordinates. GCPs are identified in aerial photos during processing to anchor the 3D model in real-world space. The more GCPs used, the higher the absolute accuracy of the final model.

The physical distance on the ground represented by one pixel in an aerial image, usually expressed in cm/px. A GSD of 2 cm/px means each pixel covers a 2x2 cm area. Lower GSD values mean higher resolution. GSD is determined by altitude, sensor size, and focal length.

A remote sensing technique that captures many narrow, contiguous spectral bands across the electromagnetic spectrum. Hyperspectral sensors can identify specific materials and compounds by their spectral signature, useful for mineral exploration and environmental analysis.

The percentage by which consecutive aerial photos overlap, both forward (along the flight line) and side (between adjacent flight lines). Photogrammetry typically needs 75-85% forward overlap and 65-70% side overlap for reliable 3D reconstruction.

The process of establishing the geometric relationship between the camera sensor and the image, including focal length, principal point, and lens distortion. Interior orientation is the first step in photogrammetric processing.

A GPS/GNSS positioning method that uses carrier phase measurements from a base station and a rover to achieve centimeter-level accuracy in real time. RTK (real-time) and PPK (post-processed) are the two kinematic methods used in drone surveying.

Near-real-time airspace approval obtained through the FAA's Low Altitude Authorization and Notification Capability. LAANC allows Part 107 operators to fly in controlled airspace near airports without waiting days for manual authorization.

Light Detection and Ranging. A remote sensing method that uses laser pulses to measure distances to the ground. Drone-mounted LiDAR can penetrate vegetation canopy to produce bare-earth models, making it essential for surveying forested or heavily vegetated terrain where photogrammetry fails.

A site-specific coordinate system that is not tied to a global or national reference frame. Local coordinates are sometimes used on construction sites for convenience but must be tied to known control points to integrate with broader survey data.

A physical shutter mechanism in a camera that opens and closes to expose the sensor, as opposed to an electronic shutter. Mechanical shutters eliminate rolling shutter distortion, which is critical for accurate photogrammetry with moving drones.

A 3D surface representation composed of vertices, edges, and faces (typically triangles). Meshes are generated from point clouds and used for visualization, measurements, and export to CAD or BIM software.

A remote sensing technique that captures images in multiple discrete spectral bands, including visible and near-infrared light. Multispectral sensors on drones are used for crop health analysis, NDVI mapping, and environmental monitoring.

An aerial photo taken with the camera pointing straight down (90 degrees below horizontal). Nadir images are the standard capture angle for photogrammetric mapping. Oblique images, taken at an angle, are sometimes added for improved 3D reconstruction of vertical surfaces.

A widely used vegetation health index calculated from near-infrared and red light reflectance. NDVI values range from -1 to 1, with healthy vegetation typically scoring 0.6 to 0.9. Drone-mounted multispectral sensors produce NDVI maps for precision agriculture.

Aerial photos captured at an angle rather than straight down. Oblique images improve 3D reconstruction of building facades, cliff faces, and other vertical surfaces that nadir-only missions miss.

A geometrically corrected, georeferenced aerial image created by stitching many individual photos into a single seamless map. Unlike a raw aerial photo, an orthomosaic has uniform scale throughout, meaning measurements taken on it are accurate like a map.

The amount by which adjacent aerial photos share the same ground area. High overlap (75-85% forward, 65-70% side) is required for photogrammetry so that software can match features across multiple images to build accurate 3D models.

A DJI drone designed specifically for mapping, featuring an integrated RTK module for centimeter-level positioning without ground control points. The P4 RTK is popular for survey-grade photogrammetry due to its accuracy and relatively low cost.

The science of making reliable measurements from photographs. In drone surveying, photogrammetry software processes hundreds of overlapping aerial images to produce orthomosaics, 3D models, point clouds, and digital elevation models.

A professional photogrammetry software suite used to process drone imagery into maps, 3D models, and survey deliverables. Pix4D is known for its survey-grade accuracy and is widely used in construction, mining, and agriculture.

A set of data points in 3D space representing the external surface of an object or terrain. Point clouds are the raw output of LiDAR and photogrammetry processing. Each point has X, Y, Z coordinates and may include color, intensity, or classification data.

The number of LiDAR points per unit area, typically expressed as points per square meter. Higher point density produces more detailed models but requires slower flight speeds or lower altitudes. Survey-grade LiDAR typically needs 50-200 points per square meter.

A GPS positioning method where raw satellite data is logged during the flight and corrected afterward using base station data. PPK often achieves slightly higher accuracy than RTK because it can use more sophisticated post-processing algorithms.

Both achieve centimeter-level accuracy. RTK applies corrections in real time during flight, which is convenient but requires a live data link. PPK logs data during flight and applies corrections afterward, offering slightly higher reliability and accuracy because it is not dependent on a stable radio link during the mission.

A GPS/GNSS positioning technique that provides centimeter-level accuracy in real time by applying corrections from a base station or network during the flight. RTK enables direct georeferencing, reducing or eliminating the need for ground control points.

The accuracy of measurements between points within a 3D model, independent of their absolute position in the real world. A model can have high relative accuracy (internal dimensions are correct) but low absolute accuracy (the whole model is shifted from its true position).

The FAA certification required to operate a drone commercially in the United States under Part 107. To obtain the certificate, a pilot must pass the initial aeronautical knowledge test and recurrent training every 24 calendar months.

The level of detail in an aerial image or derived product, typically measured as ground sample distance (GSD) in cm/px. Lower GSD means higher resolution. For topographic surveys, 2-5 cm/px is standard. For inspections, sub-cm GSD may be required.

In LiDAR, a single laser pulse can produce multiple returns as it reflects off different surfaces (canopy, understory, ground). The number and distribution of returns indicate vegetation density and structure, which is valuable for forestry and biomass estimation.

A camera shutter method where the sensor is exposed line by line rather than all at once. Rolling shutter can cause geometric distortion (jello effect) in images taken from a moving drone, which degrades photogrammetric accuracy. Mechanical or global shutters avoid this problem.

The percentage of image overlap between adjacent flight lines (parallel flight paths). Standard photogrammetry requires 65-70% side overlap to ensure features appear in multiple image rows for robust 3D reconstruction.

The photogrammetric technique of estimating 3D structure from a sequence of 2D images taken from different positions. SfM is the core algorithm behind modern drone mapping software, automatically matching features across overlapping photos to build 3D models.

The width of ground area covered by a single flight line, determined by altitude, sensor field of view, and image overlap. Wider swaths mean fewer flight lines to cover a site, but higher altitude means lower resolution.

An image applied to the surface of a 3D mesh to give it realistic appearance. In drone surveying, the orthomosaic is draped over the 3D mesh to create a photorealistic 3D model of the surveyed area.

A remote sensing technique that captures infrared radiation to visualize temperature differences. Drone-mounted thermal sensors are used for roof inspections, solar panel diagnostics, electrical inspections, and animal surveys.

Features in overlapping aerial images that photogrammetry software automatically matches to connect the images and build a 3D model. Tie points are distinct from ground control points, which have known real-world coordinates.

A survey that maps the elevation and features of a land surface, including natural terrain and man-made structures. Drone-based topographic surveys produce contour maps, DEMs, and 3D terrain models for construction, engineering, and land development.

The Universal Transverse Mercator coordinate system, a global grid that divides Earth into 60 zones. UTM is commonly used in drone surveying because it provides distance in meters, making area and volume calculations straightforward.

The FAA requirement that a remote pilot can see the drone with unaided vision (no binoculars) at all times during flight. VLOS is a core Part 107 rule. Flying beyond VLOS requires a waiver from the FAA.

The process of measuring the volume of a 3D object or surface feature from drone survey data. Common applications include stockpile inventory, cut/fill analysis for earthwork, and pit volume tracking in mining.

Open Drone Map, a free, open-source photogrammetry toolkit for processing aerial images into orthomosaics, point clouds, and 3D models. WebODM is popular among operators who want full control over processing without subscription fees.

A VTOL (vertical takeoff and landing) mapping drone that combines quadcopter hover capability with fixed-wing efficiency. The WingtraOne can cover large areas (up to 290 hectares per flight) while taking off and landing in tight spaces.

Wingtra's cloud-based post-processing software that converts raw aerial imagery into PPK-corrected, georeferenced data ready for photogrammetric processing in Pix4D, Metashape, or other software.

The rotation of the drone around its vertical axis, controlling the direction the camera faces. In automated mapping missions, yaw is managed by the flight controller to ensure the camera consistently points along the flight line.

Altitude above the home takeoff point, also called home-point altitude or relative altitude. ZTM is used by flight planning software to maintain consistent altitude over rolling terrain, though it does not account for elevation changes across the survey area.

Operating a drone beyond the pilot's unaided visual range. BVLOS operations require a special FAA waiver under Part 107 and are typically used for long-distance infrastructure inspections, pipeline surveys, and large-area mapping. Most standard drone surveys are flown within VLOS.

A 2D map representation of terrain elevation using lines of constant elevation (contour lines). Closely spaced lines indicate steep slopes; widely spaced lines indicate gentle terrain. Contour maps are a standard deliverable for topographic surveys and are used for site planning, grading design, and drainage analysis.

The process of comparing two elevation surfaces (existing ground vs proposed design) to calculate how much material needs to be cut (removed) and filled (added) on a construction site. Drone surveys produce the existing-ground surface, and CAD software computes the cut/fill volumes against the design surface.

Drone surveys are 50-70% cheaper and 10-50x faster than traditional ground surveys for sites over 5 acres. A drone can cover 50-100 acres per hour versus 1-2 acres per hour on foot. However, traditional surveys still excel at boundary determination, legal property corners, and high-accuracy control networks. Many projects combine both: drone for topographic data, traditional surveyor for legal boundaries.

Metadata embedded in each aerial photo by the drone's camera, including GPS coordinates, timestamp, altitude, camera settings, and orientation. Photogrammetry software reads EXIF data to position each image in 3D space before matching features across overlapping photos.

The process of assigning GPS coordinates to each aerial photo, either during capture (via the drone's GPS) or during post-processing (via PPK). Geotagged images enable direct georeferencing, reducing or eliminating the need for ground control points.

The two main methods for generating 3D models from drone data. Photogrammetry uses overlapping RGB photos and is cheaper, higher visual detail, and works well for open terrain. LiDAR uses laser pulses and costs 2-3x more but penetrates vegetation canopy to produce bare-ground models. Choose photogrammetry for open sites, construction monitoring, and inspections. Choose LiDAR for forested terrain, vegetation-heavy sites, or when bare-earth accuracy is critical.

The process of correcting an aerial image for lens distortion, camera tilt, and terrain relief so that it has uniform scale throughout, like a map. Orthorectification is what turns raw aerial photos into a measurable orthomosaic where every pixel represents the same ground distance.

The process of categorizing each point in a LiDAR point cloud by surface type: ground, vegetation, buildings, water, or noise. Classification is essential for producing bare-earth models (removing vegetation points) and for extracting building footprints or tree heights.

Both achieve centimeter-level accuracy. RTK applies corrections in real time during flight, which is convenient but requires a live data link. PPK logs data during flight and applies corrections afterward, offering slightly higher reliability and accuracy because it is not dependent on a stable radio link during the mission. Most survey-grade drone operators offer both.

A drone survey specifically conducted to measure the volume of material stockpiles (sand, gravel, aggregate, topsoil, ore). The drone captures aerial data, software generates a 3D surface model, and volume calculations are performed for each pile. Results are typically reported in cubic yards or tons, with material density applied for weight estimates.

Positional accuracy of 1-3 centimeters, sufficient for engineering design, construction staking, and regulatory submissions. Survey-grade accuracy requires RTK or PPK positioning, ground control points, and proper processing. Standard drone surveys without RTK/GCPs achieve 2-5 meters accuracy (mapping-grade), suitable for visualization but not measurement.

A 3D surface model made of irregularly spaced triangles connecting known elevation points. TINs are the standard format for terrain representation in CAD and GIS software. Drone survey data is often exported as a TIN for integration with engineering design tools like Civil 3D.

The accuracy of elevation values in a drone survey model, typically 1-3 times the ground sample distance (GSD). For example, at 3 cm/px GSD, vertical accuracy is approximately 3-9 cm. Vertical accuracy is critical for cut/fill calculations, drainage design, and volume estimation. It is verified using independent checkpoints.