ACTAR Photogrammetry 2 — Questions and Answers
Question 1: In close-range photogrammetry for accident reconstruction, what is the primary purpose of using coded targets or control points?
- To improve image contrast for printing
- To establish a known coordinate system and scale the model accurately (Correct answer)
- To mark the position of witnesses at the scene
- To indicate areas of vehicle damage
Correct answer: To establish a known coordinate system and scale the model accurately
Coded targets or control points provide reference locations with known real-world coordinates, allowing the software to accurately scale and orient the photogrammetric model.
Question 2: What does 'convergent photography' mean in the context of photogrammetric documentation?
- Taking photos from a single fixed position
- Capturing multiple overlapping images aimed toward a common subject from different angles (Correct answer)
- Using a telephoto lens to zoom in on distant evidence
- Photographing only the area of maximum damage
Correct answer: Capturing multiple overlapping images aimed toward a common subject from different angles
Convergent photography involves taking multiple images from different positions all directed toward the subject, providing the angular diversity needed for 3D reconstruction.
Question 3: Which type of lens distortion most significantly affects photogrammetric accuracy and must be accounted for during camera calibration?
- Chromatic aberration
- Radial (barrel or pincushion) distortion (Correct answer)
- Vignetting
- Bokeh
Correct answer: Radial (barrel or pincushion) distortion
Radial distortion causes straight lines to appear curved and introduces systematic errors in photogrammetric measurements if not corrected through calibration.
Question 4: When performing structure-from-motion (SfM) photogrammetry at a crash scene, what minimum overlap is generally recommended between consecutive images?
- 10–20%
- 30–40%
- 60–80% (Correct answer)
- 90–100%
Correct answer: 60–80%
At least 60–80% overlap between consecutive images is recommended to ensure sufficient common feature points for reliable 3D reconstruction.
Question 5: A photogrammetrist notices the point cloud of a crash scene has a significant 'dome' or 'bowl' warping artifact. What is the most likely cause?
- Using too many images
- Failure to include oblique or angled images, relying only on nadir (downward-looking) shots (Correct answer)
- Scene lighting that was too bright
- Using targets that were too small
Correct answer: Failure to include oblique or angled images, relying only on nadir (downward-looking) shots
Doming artifacts arise when all images are captured at the same height and angle (nadir-only); adding oblique images at the scene periphery corrects this systematic error.
Question 6: What is the relationship between ground sampling distance (GSD) and the height at which a UAV captures images for photogrammetry?
- GSD decreases as flight altitude increases
- GSD increases as flight altitude increases, meaning lower resolution (Correct answer)
- GSD remains constant regardless of altitude
- GSD is only relevant for satellite imagery, not UAVs
Correct answer: GSD increases as flight altitude increases, meaning lower resolution
As UAV altitude increases, each pixel covers a larger ground area, increasing GSD and reducing the spatial resolution of the final model.
Question 7: In photogrammetry software, what does the term 'dense point cloud' refer to compared to a 'sparse point cloud'?
- A point cloud captured in a densely populated urban area
- A point cloud generated using only a few high-resolution images
- A point cloud containing millions of closely spaced 3D points derived from all image pairs, compared to fewer matched keypoints in a sparse cloud (Correct answer)
- A point cloud exported in a compressed file format
Correct answer: A point cloud containing millions of closely spaced 3D points derived from all image pairs, compared to fewer matched keypoints in a sparse cloud
A dense point cloud is produced by matching pixels across many image pairs to create detailed 3D geometry, while a sparse cloud contains only the initial keypoint matches used to determine camera positions.
ACTAR Photogrammetry 2
In close-range photogrammetry for accident reconstruction, what is the primary purpose of using coded targets or control points?