Photogrammetry vs. LiDAR: Which Mapping Technology Fits Your Project?

Photogrammetry and LiDAR are the two main ways to turn a drone flight into a 3D point cloud and a survey-grade model. Both can reach accuracy of just a few centimeters, but they work on completely different principles, and each has an area where it simply outperforms the other without question.

Photogrammetry: Building 3D Models From Photos

Photogrammetry is based on the Structure from Motion principle: the drone captures a series of overlapping photos from slightly different angles, and processing software (such as Pix4Dmapper, DroneDeploy or Agisoft Metashape) identifies shared points between images and calculates depth and 3D position from them. For the reconstruction to succeed, high overlap between images is required, typically 75 to 85 percent forward overlap and 60 to 75 percent side overlap, and even more in tall structures or urban areas to avoid "blind spots" behind obstacles.

The final resolution is determined by GSD, short for Ground Sampling Distance: how much ground area is represented by a single pixel in the image. GSD depends on flight altitude, camera sensor size and focal length, so cutting flight altitude roughly in half doubles the resolution, but also doubles the number of required images and the flight and processing time. Under good conditions, with ground control points, photogrammetry achieves accuracy of 1 to 3 centimeters in open terrain.

LiDAR: Measuring Distance, Not Capturing Images

LiDAR is an active sensor: it fires laser pulses and measures their return time to calculate a direct distance to every point the pulse hits. Some pulses return from multiple surfaces in a single shot, for example from the edge of foliage and then from the ground beneath it, which is why LiDAR systems support "multiple returns," where the first return describes the upper surface and the last return reveals the ground itself beneath the vegetation.

Vegetation Penetration: The Decisive Difference

This is the difference that determines entire projects. LiDAR can penetrate gaps in vegetation and measure the ground surface directly even under significant vegetation cover, with various studies reporting penetration under as much as roughly 90 percent tree canopy coverage. Photogrammetry, by contrast, is limited to what the camera can actually see, meaning only the upper canopy surface, with effective penetration far lower, around 60 percent or less. For any project requiring a bare-earth digital terrain model (DTM) in a forest, wetland or under trees, LiDAR is almost always the only option that works.

Point Density: The Real Numbers

Professional LiDAR systems easily generate over 300 points per square meter. A concrete example is the

DJI Zenmuse L2 sensor, capable of collecting up to 240,000 points per second in single-return mode, and up to 1.2 million points per second in multiple-return mode, with up to 5 returns per pulse for better vegetation penetration. Its horizontal accuracy is 5 centimeters and vertical accuracy is 4 centimeters at a distance of 150 meters, with a detection range of up to 450 meters and coverage of up to 2.5 square kilometers in a single flight. Photogrammetry doesn't measure "point density" in the same way, since it's derived from GSD and overlap instead, but it can offer richer visual texture in open terrain., שמסוגל לאסוף עד 240,000 נקודות בשנייה בהחזר יחיד, ועד 1.2 מיליון נקודות בשנייה במצב החזרים מרובים, עם עד 5 החזרים לכל פולס לצורך חדירה טובה יותר לצמחייה. דיוק אופקי שלו עומד על 5 סנטימטרים ואנכי על 4 סנטימטרים במרחק של 150 מטר, עם טווח גילוי של עד 450 מטר וכיסוי של עד 2.5 קילומטר רבוע בטיסה בודדת. פוטוגרמטריה לא מודדת "צפיפות נקודות" באותו אופן, אלא נגזרת מה-GSD והחפיפה, אבל יכולה להציע טקסטורה ויזואלית עשירה יותר בשטח פתוח.

Cost and Equipment

A professional LiDAR sensor is a serious equipment investment, on the order of tens of thousands of dollars, and also requires dedicated processing software to classify the point cloud and filter ground from vegetation. Photogrammetry, by contrast, basically only requires a drone with a quality camera and image-processing software, making it far more accessible in terms of entry cost.

So What Should You Choose?

• LiDAR: forest inventory and canopy height models, mapping flood zones and wetlands beneath vegetation, mapping infrastructure corridors (power lines, roads, railways) that require both bare ground and vegetation-penetration analysis, and any project requiring an accurate DTM beneath trees.
• Photogrammetry: topographic surveys in open terrain, construction-progress tracking and volume calculations for mining stockpiles, cadastral mapping and general orthophoto work, and projects with a limited equipment budget.

Combining Both

Many professional mapping companies don't choose one over the other, but combine them: using LiDAR to obtain an accurate ground and elevation model, and photogrammetry images to "paint" that model with realistic visual texture for the final orthophoto. This delivers both reliable geometric accuracy beneath vegetation and a visual output that looks like a real-world image.