Unit
- Definition: Photogrammetry is the science of obtaining reliable measurements from photographs. It derives 3D coordinates, distances, and shapes by analyzing overlapping images captured from different viewpoints, without requiring physical contact with the object.
- Historical evolution: The field began in the mid-19th century with aerial balloon photography and evolved through analog, analytical, and digital phases. Modern photogrammetry leverages computer vision algorithms and high-resolution digital cameras for automated processing.
- Core principle: Triangulation uses the intersection of rays from at least two camera positions to determine the 3D location of a point. Accurate camera calibration and image orientation are prerequisites for reliable triangulation.
- Types of photogrammetry: Aerial photogrammetry uses airborne platforms like drones or aircraft for large-area mapping. Terrestrial photogrammetry employs ground-based cameras for close-range objects. The choice depends on scale, accuracy, and access.
- Comparison with LiDAR: Photogrammetry uses passive sensors (cameras capturing reflected light) while LiDAR is active (emitting laser pulses). Photogrammetry provides dense texture and color but can struggle with homogeneous surfaces and low-light conditions.
- Outputs produced: Common deliverables include point clouds, 3D meshes, orthophotos (geometrically corrected images), digital elevation models (DEMs), and contour maps. Each product serves specific analysis and visualization needs.
- Applications in industry: Surveying, construction monitoring, cultural heritage documentation, reverse engineering, forensic analysis, and environmental science. The versatility drives rapid adoption across sectors seeking accurate 3D data.
- Relation to computer vision: Structure from Motion (SfM) is a computer vision technique closely related to photogrammetry that automates orientation and sparse reconstruction. SfM is now integral to modern photogrammetry software.
Design Principle
Principle: Triangulation as the Fundamental Measurement Principle
Definition: Triangulation determines a point's 3D position by intersecting rays from two or more camera positions with known orientations. The accuracy depends on the baseline distance between cameras and the precision of angle measurements.
Application: In a typical drone mapping mission, adjacent images are captured with 60-80% forward overlap and 30-60% side overlap. This ensures every ground point appears in at least three images, enabling robust triangulation. The software automatically matches features across images to compute 3D coordinates for millions of points.
Quick Revision
- Photogrammetry extracts 3D measurements from overlapping 2D images through triangulation.
- The field has evolved from analog to digital, now incorporating computer vision algorithms.
- Triangulation requires known camera positions and orientations, achieved through calibration and orientation procedures.
- Outputs include point clouds, meshes, orthophotos, and DEMs, used in various industries.