Location/Site
Encinitas Community Park NE ball fields parking lot
(33°01’53.10″N, 117°16’45.08″W)
Date
June 29, 2026
Time
7:10 AM
Executive Summary
Mission Objective
Develop a spatial dataset of a municipal parking lot using nadir-only photogrammetry with two perpendicular lawnmower patterns to evaluate how a 90º crosshatch capture strategy affects geometry, relative accuracy, DSM quality, visual accuracy, and image overlap.
The mission aimed to assess orthomosaic, dense point cloud, textured mesh, and digital surface model outputs while continuing to document the strengths and limitations of consumer‑grade GNSS positioning. Mission 1C will compare model reconstruction to that of Mission 1A, with key data outputs including relative accuracy, warping, elongation, shifting, and 3D model detail.
Intended Applications
Potential applications include:
- Site documentation
- Relative dimensional verification
- Pavement, asset, and vegetation inventory
- Existing conditions documentation
- Baseline comparison for future site monitoring
- Capture strategy evaluation
Mission Planning
Airspace
- Class G, Uncontrolled
- No TFRs
Weather Conditions
- Temperature: 64.4º F
- Wind: 6.8 mph, 214º SSW
- Cloud Cover: BKN, 1000’ AGL
- Lighting Conditions: Diffused
Mission Success Criteria
- Complete image alignment
- Continuous orthomosaic coverage
- Minimal visible block deformation in the point cloud
- Relative measurements within ±5% of field measurements
- Clearly documented dataset limitations
- Safe and complete mission without any deviations or conflicts
Data Capture Strategy
Flight Parameters
| Parameter | Value |
|---|---|
| Flight Pattern | Automated Lawnmower Grid (manually created) |
| Altitude | 120 ft AGL |
| Flight Duration | 29m 52s |
| Camera Angle | Nadir (-90°) |
| Front Overlap | 80% |
| Side Overlap | 65% |
| Speed | 3.13 mph |
| Photo Interval | 5 sec |
| ISO | 100 |
| Shutter | 1/2000 sec |
| Aspect Ratio | 4:3 |
| Image Capture Count | 284 JPEG (8064 x 6048) |
| Lighting | Uniform Overcast |
Processing Summary
| Metric | Result |
|---|---|
| Images Reconstructed | 284/284 |
| Average Reprojection Error | 1.55 px |
| Average GSD | 0.8 cm/pixel |
| Sparse Points | 251,712 (98.7%) |
| Dense Points | 16.6 million |
| GPS Source/ CRS | Consumer GNSS (No RTK/PPK) |
| GPS Error RMS | 2.79 m |
| Outputs | Orthomosaic, DSM, DTM, Point Cloud, Mesh |
Interpretation
Orthomosaic
Produced complete visual coverage of the project area. Stitching and distortion are more pronounced in the Mission 1C model, most evident in parking stall lines, curbs, and roofing. My hypothesis is that this is due to WebODM trying to align perpendicular images with a wide field of view.
Suitable for documentation, visual inspection, asset inventory, and approximate measurements.
Point Cloud
Successfully reconstructed terrain, pavement, vegetation, and major structures. The sparse point clouds were well constructed with 99% of the reconstruction points. As expected, building facades, vehicle sides, and tall vegetation were incompletely represented due to the nadir acquisition.
Textured Mesh
Produced a coherent three-dimensional representation of the site. Quality decreased beneath mature trees and along vertical surfaces where image geometry was insufficient. Pavement melted into the vertical surfaces of vehicles.
DSM
Successfully represented overall site topography, including the gradual elevation increase across the parking lot.
Quality Validation
Relative Measurement Validation
Seven independent field measurements were compared against the reconstructed model.
Results demonstrated relative dimensional agreement generally within approximately 2–4%, with one outlier associated with a small object reconstructed from limited imagery.
Note: Actual measurements for “SE-NE Lightpost pad” and “SW Curb to White water pipes” were taken from Google Earth and may have absolute inaccuracies of 1’ to 5’.
Interpretation
| Location/object | Lat/Long Coordinates | Actual (inches) | Model (inches) | Delta (Inches) | Signed Error (%) | Model Used |
|---|---|---|---|---|---|---|
| Launch area parking stall | 33°01'53.00"N 117°16'45.06"W | 92.125 | 92.16 | 0.035 | 0.04% | Ortho |
| Slab near the building | 33°01'52.40"N 117°16'46.43"W | 123 | 120.36 | -2.64 | -2.15% | Ortho |
| North sidewalk | 33°01'55.60"N 117°16'46.51"W | 185.75 | 185.64 | -0.11 | -0.06% | Ortho |
| East sidewalk | 33°01'52.34"N 117°16'44.17"W | 89.75 | 90 | 0.25 | 0.28% | Ortho |
| Baseball table | 33°01'50.68"N 117°16'43.70"W | 35.75 | 33.72 | -2.03 | -5.68% | Ortho |
| Manhole cover | 33°01'51.21"N 117°16'44.70"W | 27 | 26.88 | -0.12 | -0.44% | Ortho |
| Dark sidewalk (SW) | 33°01'51.20"N 117°16'45.51"W | 47.5 | 47.52 | 0.02 | 0.04% | Ortho |
| SE-NE Lightpost pad | 33°01'55.4"N 117°16'46.0"W 33°01'51.2"N 117°16'42.7"W | 504 | 507.54 | 3.54 | 0.70% | Ortho |
| SW Curb to White water pipes | 33°01'53.0"N 117°16'44.3"W 33°01'52.2"N 117°16'46.8"W | 234 | 235.41 | 1.41 | 0.60% | Ortho |
| Error Calculation | Value |
|---|---|
| Mean Signed Error (Bias) | -0.91% |
| Mean Absolute Error (MAE) | 1.09 |
| Mean Absolute Percent Error (MAPE) | 1.17% |
| Root Mean Square Error (RMSE) | 1.72 |
This dataset is suitable for educational use and approximate measurements but lacks the positional accuracy required for engineering, surveying, or legal-grade geospatial analysis.
Quality Assessment
Strengths
- Complete image reconstruction
- Consistent relative measurement accuracy
- Uniform lighting enhanced detail
- Repeatable data capture workflow
Limitations
- Crosshatch data capture introduced more stitching errors and distortion
- Relative accuracy, especially in the Z-axis, had more significant GPS error
- Absolute positional accuracy limited by consumer GNSS
- Missing vertical surfaces and tree occlusions
- Vehicle movement introduced localized artifacts
- Observable large-scale block deformation on the sidewalk point cloud
Conclusion
Lessons Learned
- The crosshatch pattern improved the model’s relative accuracy, as evidenced by a reduction in RMSE across all measurement validations.
- The 90º rotation of the camera’s aspect ratio in the crosshatch pattern introduced more stitching artifacts and warping between overlapping images.
- Crosshatch data capture increased GPS error, mostly in the Z coordinate..
Final Statement
This mission successfully compared a 90º crosshatch data capture with a longitudinal lawnmower pattern. I expected the 90º crosshatch pattern to provide additional coverage, thereby reinforcing the model’s detail and geometry. The relative accuracy improved, as evidenced by the lower RMSE of 1.72, but visually, there was increased warping. This didn’t appear to affect any of the measurement validations, but it could be an issue when measuring other aspects of the model. Vertical surfaces were inadequately reproduced and thus unsuitable for measurement.
This dataset is suitable for progress status, inventory assessment, horizontal surface inspection, and approximate measurements, but would not be adequate for engineering, surveying, or construction-grade decision-making.