Free tools Windows power users keep installed
One-click scans. No signup required.
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
To clean and validate point cloud data from a laser scanner, first align the scans, remove obvious unwanted objects, then filter isolated noise on a copy of the data. After filtering, check registration, completeness, and accuracy against independent survey control; filtering alone does not prove accuracy. For building and scan-to-BIM work, follow the project’s registration and control tolerances. U.S. Geological Survey (USGS) 3DEP deliveries have separate classification, accuracy, and reporting rules, so do not treat a USGS quality level as a building-model tolerance.
What does cleaning a point cloud prove—and what does it not prove?
Cleaning makes a cloud easier to inspect, register, classify, and model by flagging or removing points that do not represent the surface of interest. It does not establish that coordinates are accurate. A cloud can be internally tight yet shifted from surveyed truth; conversely, a cloud may match control overall but contain local noise, gaps, or poor overlap.
Keep the checks distinct:
- Absolute accuracy: How closely does the data match independent surveyed checkpoints?
- Registration or relative precision: How well do scans or flight strips agree with one another?
- Completeness: Are there gaps, occlusions, edge artifacts, or duplicate coverage?
- Noise: Are there returns that do not plausibly belong to the surfaces being modeled or classified?
For terrestrial and handheld scanner projects, registration is usually assessed before filtering is trusted. For airborne or UAS lidar, relative strip alignment and absolute checkpoint accuracy are separate checks. USGS 3DEP requires accuracy verification and a detailed validation report before classification and derivative products; its processing and handling requirements make clear that removing noise is not a substitute for validation.
Popular on Amazon
- Ergonomically Designed: Work in tight areas with a compact design that gets into tough spots
- Both tools are designed to fit into difficult to reach spaces
- Complete a wide range of applications with a high-performance motor that delivers up to 300 UWO, and an impact driver delivering up to 1,400 in-lbs of torque
- Both the DEWALT impact driver and drill driver feature integrated LED work lights with a convenient 20-second delay, ensuring enhanced visibility in dimly lit or challenging work areas
- One-Handed Loading - Keep one hand free with a 1/4 in. hex chuck that accepts 1 in. bit tips
How should you align scans before filtering?
Align the scans first: a noise filter cannot repair a rotated, bowed, or misregistered scan. For terrestrial or heritage scans, a common sequence is coarse alignment using picked point pairs or known targets, followed by iterative closest point (ICP) refinement, then noise review. Treat the final ICP RMS as a registration-quality indicator, not as an absolute-accuracy result; it should be interpreted in relation to sensor noise and the project specification.
For building scans, published commercial guidance offers useful practice figures but not a national acceptance standard. The Future 3D describes registration within 3 mm between scan pairs and overall accuracy within 5 mm of survey control for BIM-targeted work. A separate scan-to-BIM checklist describes mean registration error below ±5 mm as a pass line for many architectural projects and global fit to survey control below 6 mm or the project specification. These are industry-practice statements, not USGS or ASPRS accuracy classes. If the contract sets a different tolerance, use the contract.
When registering overlapping scans, use the actual shared surface as the overlap region rather than assuming the whole scan is common. A low registration RMS can still conceal poor coverage or a systematic shift if the overlap is small, repetitive, or geometrically weak. Check the fit visually at edges, corners, and other distinctive surfaces as well as reviewing the software’s summary statistic.
How do you remove obvious objects and gross noise?
Before running a statistical filter, segment objects that are clearly not part of the desired surface: people, vehicles, scaffolding, stray equipment, sky, or background beyond the intended scene. Use a manual selection, distance fence, project boundary, or a defensible geometric limit when the unwanted region is known. This avoids asking an outlier algorithm to distinguish a large object from legitimate geometry.
For a building model, retain objects that belong in the scope even if they are isolated or unusual. A light fixture, column, cable, or narrow structural member may be a valid feature rather than noise. Conversely, moving people or vehicles can leave smeared or doubled returns. Decide the intended deliverable first: a clean architectural shell, a record of existing conditions, and a complete classified lidar dataset may require different treatment.
Visual segmentation is especially important for objects such as people and vehicles, which can form dense, coherent point clusters and therefore survive neighbor-based filters. Inspect scan edges, reflective or glass surfaces, and places where multipath or occlusion is plausible. Filters are aids to review, not a reliable substitute for understanding what the scene contains.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Which outlier filter should you use?
Choose the filter to match the shape of the noise and the cloud’s density. Statistical outlier removal (SOR) is a reasonable starting point when sampling density is fairly even and unwanted points are isolated; radius-based removal is often more suitable when density varies. Neither has a universal neighbor count or threshold.
Also popular on Amazon
- Valued Carpenter Pencil Set: You will get 2 pcs solid carpenter pencils with 26 piece 2.8 mm refills, 1 replaceable sharpener, 1 plastic storage box.The complete carpenter pencils combination allows you to finish your work faster and more easily
- Deep Hole Marker Pencil: The deep-hole construction pencils adopts 45mm elongated tip design, which is more convenient to mark in the small hole or in other tight areas that other carpenter markers cannot reach
- Carpenter Pencils with Sharpener: The sharpener is screwed into the top of the work pencil, which won't get lost either. Built-in pencil sharpener that keep the lead with pointed and smooth to Improves line of sight in fine work
- Stronger Solid Lead: This work pencil is matched with a 2.8 mm thick lead , which is much thicker and stronger during the drawing process of construction work, it will not break or damage easily
- Marks on Various Surfaces: 3 colors solid construction pencil can marks on various surfaces,such as metal, plastic, wood, paper etc. Ideals for woodworkers, contractors, craftsmen, builders, merchants and masons
| Method | What it flags | Useful when | Watch for |
|---|---|---|---|
| Statistical outlier removal (SOR) | Points whose mean distance to their k nearest neighbors is unusually large compared with the cloud’s overall distribution. | Noise consists of isolated points and density is reasonably consistent. | High thresholds may leave noise; aggressive thresholds may remove valid sparse features or returns on steep surfaces. |
| Radius outlier removal | Points with fewer than a specified number of neighbors inside a search radius. | Local neighbor counts are more meaningful than one cloud-wide distance distribution, particularly where density varies. | A radius that is too small can remove legitimate sparse geometry; one that is too large may retain isolated noise. |
| Extended-local-minimum filter | Low pits that sit within a local cluster rather than standing apart from it. | Low outliers survive ordinary neighbor-distance filters, particularly in some terrain workflows. | It is a specialized ground/terrain tool, not a general building-scan cleanup step. |
Published parameter examples vary because point spacing, scene geometry, and noise differ. Enginyring gives a SOR starting example of k = 20 and a standard-deviation multiplier of 2.0–3.0, then describes radius search at 2–3 times average point spacing with a minimum of 5–10 neighbors. Its sequence is radius first to remove isolated specks, then SOR. PDAL’s workshop demonstrates a different example: statistical mode, multiplier 3, and mean_k 8. That workshop says the filter classifies outliers as low/noise points using LAS class 7; its additional elevation expression is specific to its example file, not a general height limit.
A 2026 forestry tutorial describes PDAL defaults of mean_k 8 and multiplier 2.0, and suggests mean_k 8–16 and multiplier 2.0–2.5 in dense forest. It gives radius mode examples for variable density and suggests an extended-local-minimum pass for low pits. Those are source-specific recommendations, not settings to copy blindly into a building scan. The tutorial warns that a multiplier below 2.0 may start removing valid returns on steep slopes. Consult the documentation for the installed software version before relying on a default.
In PDAL, the denoising workshop shows how an outlier stage can classify noise before a later expression excludes it from a derived output. The workshop notes that the statistical filter loads the whole cloud into memory, so very large files may need to be split. CloudCompare, Leica Cyclone 3DR, Autodesk ReCap Pro, FARO SCENE, Terrasolid, and LP360 are also cited in current how-to material as environments for point-cloud cleaning or classification. Tool availability does not make their settings interchangeable: check the relevant version’s documentation and preserve the original data.
Recommended Free Tools
For architectural scans, a vendor guide recommends combining SOR with bilateral smoothing: SOR targets floaters, while bilateral smoothing is intended to reduce noise while retaining edges. Smoothing changes point positions or surface character, so do not apply it automatically to a survey deliverable or use it to conceal registration problems. Keep an unfiltered reference and validate any altered copy against the original scope and tolerance.
Should you classify, withhold, or delete noise?
For a modeling copy, it may be appropriate to export a filtered file that omits points outside the scope. For a USGS 3DEP delivery, preserve the audit trail: identify non-believable surface returns with the LAS 1.4 withheld bit, and use low-noise class 7 or high-noise class 18 as applicable. Deleting points is a separate, explicit export decision.
The USGS Lidar Base Specification 2025 rev. A says the withheld flag is for points that cannot reasonably be interpreted as valid surface returns, including outliers, blunders, geometrically unreliable points, aerosol backscatter, multipath, airborne objects, and sensor anomalies. It may be combined with low- or high-noise classes and should be used when those conditions apply. USGS requires proof of performance, preferably Maximum Surface Height Rasters; rebuild the proof if the point cloud changes. The overlap bit flag is not used for point clouds intended for 3DEP.
USGS minimum classes in Table 5 include class 1 (processed but unclassified), 2 (bare earth), 7 (low noise), 9 (water), 17 (bridge deck), 18 (high noise), 20 (ignored ground, typically near breaklines), 21 (snow, if present), and 22 (temporal exclusion). In a classified delivery, class 0 is not permitted unless the point is withheld. Extra classes are allowed, but USGS does not grade classes outside the minimum scheme. Inconsistent classification from tile to tile or swath to swath can be grounds to reject a delivery. Water points are checked when tied to a breakline, and exact duplicates—same x, y, z, and timestamp—as well as near-duplicates with a small consistent shift, are rejected.
Also popular on Amazon
- 【Great Compatibility】This Katerk 1/4 inch hex shank bit holder is specifically designed for 1/4 inch hex shank drill bits. It's compatible with most 1/4 fast hex handles, hex sockets, various electric screwdrivers, and handheld screwdrivers. The bit holder makes it a valuable addition for any handyman.
- 【Secure and Safe】Built with a secure backup nut design, each drill bit holder securely locks onto your bits, ensuring they stay firmly in place. Additionally, our bit holder incorporates a high-quality steel ball rolling design that holds up to several kilograms of weight, ensuring your various drill bits don't fall off.
- 【Easy One-Handed Operation】The bit holder for impact driver allows you to change bits single-handedly, simplifying your workflow. Its multi-color design further allows for quick identification of the drill bit you need.
- 【Compact and Convenient】Thanks to its compact size, this 1/4 inch bit holder is easy to carry around. The bit holder allows for easy attachment to various tools, making this a convenient addition to your construction accessories. The Katerk bit holder is cast from high-quality alloy material, promising a long product lifespan. Despite its rugged strength, the bit holder remains lightweight, making it portable.
- 【Cool Christmas Gift For Men Stocking Stuffers】 This screwdriver bit holder, driver bit holder, impact bit holder, can be given as a gift to your loved one, especially for anyone involved in construction or electrical work. It's a must-have for stocking stuffers for men and women, tools gifts for dad, tech gadgets for men, gifts for dad, gifts for him, gifts for husband, gifts for boyfriend, cool gadgets for men, and cool gifts for dad.
For a modeling copy, check that classes 7 and 18 are excluded from the ground surface or otherwise ignored by the ground filter. Keep the classified or raw source file separately so that a later reviewer can distinguish flagged points from points actually removed from an export.
How do you validate absolute accuracy with checkpoints?
Compare the point data against independent, accurately surveyed checkpoints that were not used to calibrate acquisition. Report RMSE and the checkpoint survey uncertainty; a checkpoint test is only as strong as the control used to perform it.
Under the ASPRS Edition 2 checkpoint rule described in 2024 sources, at least 30 checkpoints are needed to report “Tested to Meet” accuracy. Projects with fewer checkpoints can use “Produced to Meet”; the Edition 2 announcement also lists a cap of 120 checkpoints for large projects. Distribute checkpoints across the project and relevant land-cover types rather than clustering them in one convenient area. Check the applicable ASPRS Edition 2 Version 2 standard and addenda for contractual reporting requirements: the February 2024 journal article discussing the 30-point rule predates the board’s 24 June 2024 adoption of Version 2.
For USGS lidar, nonvegetated accuracy (NVA) checkpoints belong in clear, open, undisturbed ground that produces single returns; vegetated vertical accuracy (VVA) checkpoints belong in vegetation. The USGS process calls for daylight photographs of each checkpoint from the four cardinal directions, with the instrument visible and enough resolution to inspect at least 10 feet around the point. Compare checkpoints to a TIN made from ground-classified points. Test the DEM separately against the bare-earth surface. Report four results: NVA and VVA for the point data, and NVA and VVA for the DEM. NVA must meet the specification; VVA is reported.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Checkpoint survey error is part of the total, not a negligible footnote. The model described by LP360 Support for ASPRS 2024 reporting is Total RMSE = √(Product RMSE² + Survey RMSE²). If the checkpoint survey itself is uncertain, the product cannot honestly be claimed as more accurate than the combined result suggests. LP360’s summary of field practice describes at least two—and preferably three—independent GNSS occupations, with re-initialization from a different position (at least 15 feet horizontally and 2 feet vertically), and occupation times of at least 180 seconds for RTK or RTN and 300 seconds for RT-PPP. Treat those figures as the support article’s summary of practice, and confirm requirements against the applicable standard and project method.
LP360’s 2025 summary gives approximate survey RMSE by method: RTK GNSS, about 2 cm horizontal and 3 cm vertical; RTN, about 1–2 cm horizontal and 2–3 cm vertical; RT-PPP, about 1–2 cm horizontal and 3–5 cm vertical; static GNSS, under 1 cm horizontal and under 2 cm vertical; and total station, under 1 cm in both components when canopy requires an optical survey tied to GNSS control. These are approximate method figures from that source, not guaranteed results. Choose a survey method whose uncertainty is appropriate to the accuracy you intend to report.
What USGS quality level or RMSE target applies?
For 3DEP, quality level is a package combining pulse spacing, density, and accuracy requirements. It is not the same system as an ASPRS centimetre accuracy class, and neither is a generic scan-to-BIM tolerance. USGS Table 4 gives the nonvegetated vertical RMSE limits below; Table 1 gives the density and aggregate nominal pulse spacing. The figures are from the USGS Lidar Base Specification 2025 rev. A tables.
Also popular on Amazon
- Toolwiz 17-inch magnetic sweeper maintaining a load capacity of up to 8.8 pounds provides a perfect solution to the problem of clearing small ferrous metal objects, such as nails, tacks, bolts, and metal shavings from workshops or garage floors.
- Convenient and portable：17” is good for home, small workshop, or small yard as it is space-saving and more reachable for narrow areas. Perfect for collecting nails or small parts in the car, under the table, and in other nooks or crannies.
- Upgraded Handle Design: The handle features a sturdy, non-telescopic design to completely prevent slipping during use. By simply adding or removing the extension poles, the height can be adjusted to 3 different lengths: 18.9" (48 cm), 32.5" (82.5 cm), and 42.5" (108 cm) from ground to top. This customizable height helps prevent back bending and muscle strain, making your cleaning tasks effortless
- Flexible wheels: Upgrade wheel screws to avoid being too tight to spin or too loose. More comfortable 3-inch rubber wheels roll smoothly on carpet, grass, concrete, and hardwood floors. Just push and pull this rolling magnet-sweeper to pick up metal nails easily.
- Lightweight materials: The magnetic nail sweeper's compact and lightweight design makes the sweeping can be done anywhere you want, Push or pull it under the furniture, passing through the workshop floor, driveway, mud, grass, and gravel.
| USGS quality level | Nonvegetated vertical RMSE | Aggregate nominal pulse spacing | Minimum pulse density |
|---|---|---|---|
| QL0 | ≤ 0.050 m | ≤ 0.35 m | ≥ 8.0 pulses/m² |
| QL1 | ≤ 0.100 m | ≤ 0.35 m | ≥ 8.0 pulses/m² |
| QL2 | ≤ 0.100 m | ≤ 0.71 m | ≥ 2.0 pulses/m² |
| QL3 | ≤ 0.200 m | ≤ 1.41 m | ≥ 0.5 pulses/m² |
USGS Table 2 also sets internal precision limits. The maximum smooth-surface RMSD and swath-overlap RMSD are respectively ≤0.03 m and ≤0.04 m for QL0; ≤0.06 m and ≤0.08 m for QL1 and QL2; and ≤0.12 m and ≤0.16 m for QL3. These are not the same as checkpoint RMSE. Check the USGS tables and the exact revision in the task order before putting limits into a contract; the supplied tables page uses a 2025 rev. A header while its introductory line refers to tables referenced in the 2024 rev. A specification.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
For CONUS 3DEP work, unless agreed otherwise with USGS, the specification identifies NAD 83 (2011 realization, epoch 2010.00) for horizontal positions and ellipsoid heights, NAVD 88 for orthometric heights, and GEOID18. Alaska and island territories are agreed project by project. Horizontal and vertical units should match; metres are preferred, and mixed metres and feet are treated as a variance. The specified delivery is LAS 1.4-R15, Point Data Record Format 6, 7, 8, 9, or 10, with CRS recorded as OGC 2001 well-known text. These are 3DEP delivery conventions, not defaults to impose on a building scan without checking its specification.
How do you check internal precision and swath overlap?
Internal precision describes consistency within and between strips or scans; it does not establish absolute accuracy. A cloud can have excellent internal precision and still be offset from control. USGS smooth-surface testing uses single-return hard surfaces such as parking lots and large roofs, with sample areas of about 100 pixels across the left, center, and right of a swath.
For the USGS within-swath test, the slope-corrected cell precision is calculated as Precision = Range − (Slope × CellSize × 1.414), where Range is the highest minus lowest point in a cell, Slope is the maximum slope to the eight neighboring cells (as a decimal, measured from minimum elevation), and CellSize = CEILING(ANPS) × 2. Report a polygon of sample areas with minimum and maximum slope-corrected range and RMSD in z. The resulting RMSD must fall within the applicable Table 2 limit.
USGS also evaluates swath separation across the work unit. In localized comparisons on roads, open ground flatter than 10 degrees, and roofs flatter than 10 degrees, delta-z of 8 cm or less is acceptable; contiguous, prevalent pixels above 8 cm are not. For a broader view across terrain types, flight-line-shaped differences greater than 16 cm are unacceptable. Results between 8 and 16 cm may warrant more review rather than an automatic fail. Hydrology and snow are excluded from pass/fail because conditions can change over time, but remain visible in the images.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →ASPRS guidance described in a 2024 journal article gives a separate rule of thumb: smooth-surface precision should be no greater than 0.6 times the vertical accuracy class of the derived map product. The article also discusses a principal-component-analysis method that measures variation along the surface normal on planar cells, which can be more informative than elevation range on sloped surfaces. That is a research and software workflow described by the authors, not a method named as a USGS requirement.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How can you check density and completeness?
Do not treat a project-wide average pulse density as proof that coverage is complete. An average can hide stripes, holes, or local areas where occlusion has removed the surface of interest. Review density spatially and inspect the actual coverage of the surfaces needed for the project.
Also popular on Amazon
- APPLY TO MOST DRILLS : 6 pieces screw extractors work with any drill and all the screw size or bolt (3-12mm), These extractor tools are used to reverse the damaged screw counterclockwise until it is released
- The DIY Tools for Men-Men won't know they need screw extractor set until they have one! This screw extractor set is GIFTS WRAP AVAILABLE, can be gifted to your loved ones as it reflects a fresh trend with fun and practical. Amazing and Unique tool gifts for men women. The must have stocking stuffers for men, gifts for dad, husband, father, him, guy teens, adult, Boyfriend, son on Birthdays, Christmas, Fathers Day or anniversary. Perfect Holiday Gifts Guide
- HIGH QUALITY - This damaged screw remover tool kit is made from High-Speed Steel H.S.S 4341# , which have high hardness of 63hrc, it's durable and easily to remove any damaged screws.
- EASY TO USE: Just two simple steps to get the job done - drill a hole with a cutting size using the right size screw extractor and easily remove the screw or bolt using the removal tool Hardness 63-65HRC: Titanium hardened steel material, more durable than most screw extractors on the market, better hardness and durability than others, buy with confidence and complete peace of mind
- Cool Gadgets for Men Dad Christmas Birthday Gifts for Men and Women - Men won't know they need screw extractor sets until they have one! Every screw has the potential to rust and break, which can create a situation where it cannot be removed. Owning it, there will be no anxiety about rusty screws. This screw extractor set is GIFTS WRAP AVAILABLE, can be gifted to your loved ones as it reflects a fresh trend with fun and practical. Amazing and unique tool gifts for men women. A must have stocking stuffer for men on birthdays, Christmas, Father's Day or anniversaries, a gifts for dads, husbands, fathers, him, adults, boyfriends, sons. The best and most practical and surprising Christmas gifts
A June 2026 Geo Business session on Global Mapper v27.0 presents a Voronoi approach, attributed to ASPRS recommendations, as a more informative density check than a coarse grid, alongside tools for horizontal and vertical offsets against control. The source material does not provide a full procedure to reproduce here, so do not infer cell sizes, acceptance percentages, or pass thresholds from that reference alone. Use the method and density definition required by the project or governing specification.
For terrestrial construction scans, completeness is also a line-of-sight question. A clean floor cloud does not prove that the back of a column, the ceiling above services, or a recess behind equipment was captured. Compare the point cloud with the intended scope, review occluded regions, and identify areas that require a return visit instead of filling them with an unsupported assumption.
What targets help validate horizontal and 3D accuracy?
Use visible, surveyed targets when the project needs a defensible horizontal or three-dimensional check. High-contrast paint or alternating black and white patterns make target centers easier to identify in the point cloud. The 2024 journal article describes ground-based checkerboard or concentric targets for XY checks and spheres or discs arranged in a defined pattern for XYZ checks.
A leveled “Accuracy Star” design in that article uses six reflective discs on a hexagon. Three or more such targets can support a full translation-and-rotation solution. Checkerboards and concentric panels support an automated XYZ offset, but not that full six-degree solution; survey nails remain a vertical-only check. These are target options for validating the cloud, not scanners or replacements for independent survey control.
The same article reports one highway-intersection drone case with a TrueView 680 survey and a comparison to 21 photogrammetric panels after correction. It gives a final total RMSE(z) of 0.57 cm (0.019 feet), but the centimetre components it prints—0.33 cm and 0.5 cm—combine by root-sum-square to about 0.60 cm, not 0.57 cm. The article’s labels for the two components also conflict with its earlier definitions, and 21 panels are below the 30-checkpoint threshold for “Tested to Meet.” Treat this as a single-site reported result with caveats, never as a construction tolerance or universal benchmark.
What should the final quality-control review include?
Before delivery, check the cloud as a whole as well as the filter’s output. A practical closeout should document both what was changed and what evidence supports acceptance.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problems- Source and edits: Preserve the raw file; record filters, settings, software/version, and whether unwanted points were flagged or deleted.
- Registration: Review pairwise fit, overall fit, scan or strip edges, and the project’s control comparison. Do not substitute registration RMS for checkpoint accuracy.
- Noise and classification: Inspect for floating ground points, high residual noise, low pits, moving-object smear, and plausible vertical range. For a modeling copy, confirm noise classes are excluded from the intended surface.
- Completeness: Inspect coverage gaps, occlusions, duplicates, edge artifacts, and density distribution. Verify that every required area is represented.
- Surface and materials: Check for glass or mirror multipath, anomalous reflections, and artifacts near edges or reflective surfaces.
- Deliverable integrity: Confirm tiles edge-match, the agreed CRS and units are present, and any required geoid metadata names the correct file.
- Evidence: Retain checkpoint statistics and survey uncertainty, internal-precision and overlap checks where required, density/completeness review, and the project’s stated acceptance criteria.
For USGS work, do not classify or create derivatives before the required point-data accuracy verification. Follow the 2025 rev. A specification and the revision actually named in the task order, and rebuild required proof of performance after changing the cloud. For a building scan, follow the registration and survey-control tolerances in the project specification; a USGS quality level is not a scan-to-BIM acceptance class.
Frequently Asked Questions
Can I make a clean copy without losing the raw scan?
Yes. Keep the original immutable and write filtered or classified results to a separate output. Record which points were flagged, omitted, or smoothed so another person can distinguish a reversible classification decision from a changed export.
Does a low ICP RMS mean the building is in the right place?
No. ICP measures how well overlapping geometry fits, not whether the registered cloud matches an independent coordinate reference. A common offset can remain even when scans agree closely with one another.
Why can a USGS lidar file have noise points that are not deleted?
The withheld bit and noise classes let a 3DEP delivery identify questionable returns while retaining an auditable point set. A downstream modeling export can omit those points, but that is a separate output decision.
