Digital Photogrammetry & Point Clouds Flashcards

What is the difference between a DSM and a DTM?

Answer: DSM (Digital Surface Model) includes all features above ground; DTM (Digital Terrain Model) represents only bare earth

A DSM (Digital Surface Model) represents the top of all surfaces including buildings and vegetation, while a DTM (Digital Terrain Model) represents only the bare-earth terrain after removing above-ground features.

What is 'occlusion' in the context of dense image matching?

Answer: Areas not visible in one or more images due to blocking by objects

Occlusion occurs when objects (buildings, trees) block the line of sight to certain areas in some images, creating gaps or reduced accuracy in the resulting point cloud.

In UAV/drone photogrammetry, what overlap configuration is typically recommended for high-quality 3D reconstruction?

Answer: 80–90% forward overlap and 70–80% sidelap

High-quality UAV 3D reconstruction typically requires 80–90% forward overlap and 70–80% sidelap to ensure every surface area is seen from multiple angles for robust matching.

What does the LAS file format store in photogrammetric and LiDAR workflows?

Answer: Standardized binary point cloud data including XYZ coordinates, intensity, classification, and return information

The LAS format is an ASPRS-standardized binary file format for storing point cloud data, including XYZ coordinates, intensity, RGB color, classification codes, GPS time, and return numbers.

What is 'noise filtering' in point cloud processing?

Answer: Eliminating outlier points that do not represent actual surfaces (e.g., flying points)

Noise filtering removes outlier or erroneous points—such as isolated flying points far from the main surface—that would degrade the accuracy of derived terrain models.

What is 'multi-ray' matching in photogrammetric dense reconstruction?

Answer: Simultaneously using three or more overlapping images to improve matching reliability and accuracy

Multi-ray matching uses three or more overlapping images simultaneously to verify and improve the reliability of matched points, reducing errors compared to stereopair-only matching.