ASPRS accuracy (Edition 2)
Horizontal and vertical accuracy by the Positional Accuracy Standards, Edition 2: product accuracy with the checkpoints' own error, against its class, as found, blunders, mean error, and the 30-checkpoint minimum.
Planning and education aid. Not for primary navigation. Full disclaimer
A published result changed on 2026-09-24: ASPRS accuracy (version 1.1.0) is stable and now takes a list of checkpoint errors: it computes the RMSEs, tests NVA and horizontal accuracy against… Changelog
Vertical accuracy is 2.24 cm RMSE. The checkpoint count meets the Edition 2 minimum of 30.
- Checkpoint check
Provenance
- Computed by
- drone.photogrammetry.asprs-accuracy 1.1.0, core 0.1.0
- Model
- ASPRS Edition 2: RMSE = √(Σe² / n) per component; RMSE_H = √(RMSE_x² + RMSE_y²); product accuracy = √(RMSE_fit² + RMSE_checkpoint²) per component, so horizontally √(RMSE_H1² + RMSE_H2²) with RMSE_H2 = √2 × the per-axis survey RMSE (section 7.11); NVA from open-ground checkpoints must be at or under its class, VVA from vegetated ones is reported; mean error under 25% and any error over 3 times the target RMSE per component (horizontal target ÷ √2 per axis) a blunder (7.2); checkpoints at least twice as accurate as the product (7.12) and at least 30 (7.13)
- Accuracy
- Exact to the standard's definitions
- Notes
- None
- Cites
- American Society for Photogrammetry and Remote Sensing, ASPRS Positional Accuracy Standards for Digital Geospatial Data, Edition 2
Something look off?
How we got thisFormula, worked example, sources, and proof
Model: ASPRS Edition 2: RMSE = √(Σe² / n) per component; RMSE_H = √(RMSE_x² + RMSE_y²); product accuracy = √(RMSE_fit² + RMSE_checkpoint²) per component, so horizontally √(RMSE_H1² + RMSE_H2²) with RMSE_H2 = √2 × the per-axis survey RMSE (section 7.11); NVA from open-ground checkpoints must be at or under its class, VVA from vegetated ones is reported; mean error under 25% and any error over 3 times the target RMSE per component (horizontal target ÷ √2 per axis) a blunder (7.2); checkpoints at least twice as accurate as the product (7.12) and at least 30 (7.13)
Accuracy: Exact to the standard's definitions
When to use this: Use this to write the accuracy statement for a mapping product by the current ASPRS standard: an orthomosaic, a drone survey, a lidar surface, or a planimetric map checked against surveyed checkpoints. Give the fit RMSEs you already have, or paste the checkpoint errors and let it compute them, and it gives the product accuracy with the checkpoints' own error folded in, whether the product meets its horizontal and NVA classes, the VVA as found, and the checks the standard asks for before a statement can be made: enough checkpoints, accurate enough, no blunders, and no bias.
Limitations: Accuracy here is RMSE only, as Edition 2 requires; the 95% confidence figures of older standards are not computed, and a 95% number should not be quoted as an ASPRS accuracy. The checkpoint survey RMSE is taken per axis, the same in easting, northing, and height, as the standard assumes when it combines them; if the surveyor reports a horizontal radial figure, divide it by √2 first. The standard applies its blunder limit per component but states horizontal classes as RMSE_H, so the per-axis horizontal target is taken as the class divided by √2. Blunders are listed for investigation, never removed, and vegetated height errors are not tested against a limit, since VVA has no pass or fail. It cannot tell whether checkpoints are well defined, well distributed, or on slopes under 10%, which the standard also requires.
Worked example: 1.00 cm fit with 2.0 cm checkpoints. Source: ASPRS Positional Accuracy Standards, Edition 2, Table 7.4, first row: 1.00 cm fit and 2.0 cm survey checkpoints give 2.24 cm. It is golden test vector v001, and every build checks the tool still gives its answer within its tolerance.
You enter
- Checkpoint survey RMSE
- 2.0 cm
- Number of checkpoints
- 30
- RMSE z
- 1.00 cm
You get
- Vertical accuracy (RMSE_V)
- 2.24 cm
- Checkpoint check
- The checkpoint count meets the Edition 2 minimum of 30.
Review: Not yet independently reviewed by a Part 107 remote pilot.
Last verified: 2026-09-25, when a maintainer last confirmed this tool's sources at the issuer. See the sources ledger.
Status: version 1.1.0, core 0.1.0. See this tool in the verification report.
Changes
- 2026-09-24, result change: ASPRS accuracy (version 1.1.0) is stable and now takes a list of checkpoint errors: it computes the RMSEs, tests NVA and horizontal accuracy against their classes, reports VVA as found, lists blunders over three times the target RMSE, and flags a mean error of 25% of the target or more and checkpoints not twice as accurate as the product. Answers for the old inputs are unchanged. Promotion removed EXPERIMENTAL_TOOL from its warnings, which moved INSUFFICIENT_CHECKPOINTS from index 1 to index 0; the superseded vector pinned the old index. Changelog
Checked against: 34 golden test vectors (download the test vectors, each with its source and tolerance). See how results are checked and every source.
Constants this tool assumes
Values you do not enter, and where each comes from.
- Minimum number of checkpoints
30 checkpoints— ASPRS Positional Accuracy Standards for Digital Geospatial Data- Maximum absolute mean error as a share of target RMSE
25 %— ASPRS Positional Accuracy Standards for Digital Geospatial Data- Component blunder threshold
3 times target RMSE— ASPRS Positional Accuracy Standards for Digital Geospatial Data- Required checkpoint-to-product accuracy ratio
2— ASPRS Positional Accuracy Standards for Digital Geospatial Data
Sources
- ASPRS Positional Accuracy Standards for Digital Geospatial Data, Edition 2, American Society for Photogrammetry and Remote Sensing, Edition 2 (Version 1.0, February 2023, as read; Version 2.0 was approved in 2024). Section 7.2 (mean error and blunders), 7.3 and 7.4 (horizontal and vertical accuracy, NVA and VVA), 7.11 (product accuracy with checkpoint error, Table 7.4), 7.12 and 7.13 (checkpoint accuracy and the 30-checkpoint minimum).
Terms
- ASPRS: American Society for Photogrammetry and Remote Sensing
- The society that publishes the Positional Accuracy Standards for Digital Geospatial Data used to state map accuracy. Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
- NVA: non-vegetated vertical accuracy
- The height accuracy of an elevation product measured at checkpoints on open ground, such as bare soil, short grass, or pavement. It must meet the product's accuracy class. Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
- RMSE: root-mean-square error
- The square root of the average squared difference between measured and true values. ASPRS accuracy classes are stated with it. Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
- VVA: vegetated vertical accuracy
- The height accuracy of an elevation product measured at checkpoints in vegetation, such as tall weeds, crops, brush, or forest. It is reported as found, with no pass or fail. Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
ASPRS — American Society for Photogrammetry and Remote Sensing
The society that publishes the Positional Accuracy Standards for Digital Geospatial Data used to state map accuracy.
Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
NVA — non-vegetated vertical accuracy
The height accuracy of an elevation product measured at checkpoints on open ground, such as bare soil, short grass, or pavement. It must meet the product's accuracy class.
Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
VVA — vegetated vertical accuracy
The height accuracy of an elevation product measured at checkpoints in vegetation, such as tall weeds, crops, brush, or forest. It is reported as found, with no pass or fail.
Source: ASPRS Positional Accuracy Standards for Digital Geospatial Data
Learn the concept: What is ground sample distance (GSD)?