Grass over the edge, not canopy — veg-notch v2 (visual re-check)

2026-08-01 · AI3D-337 · re-adjudication of the same 65 sampled notches (population 2825), this time judging only the physical scene — detected-edge polylines explicitly ignored as unreliable · supersedes v1 report

TL;DR. Re-examined with the detector taken out of the loop, the picture changes: ~66% of the sampled notches are detector artifacts — the edge polyline excursions have no physical counterpart at all. No real notches in the asphalt were confirmed anywhere (0 of 65): every decisive cross-section shows the pavement ending at the same lateral position inside and outside the "notch", smooth and unbroken. Vegetation is involved at only ~15% of sites, and where the point clouds are decisive it is exactly what Miro's sketch suspects: grass from the verge leaning over an intact, straight pavement edge — scale 0.1–0.3 m — never tree canopy or hedges over the roadway (the one hedge case abuts the edge without overhanging it). Grass overhang is real but small; it cannot explain 0.3–1 m notches, though it does trigger them indirectly: blades + raised verge occlude the pavement rim, the data thins, and the detector wanders inboard.

Verdict on the two sketches

Miro's sketch poses the actual question: left — a genuine notch in the pavement with grass in it; right — a straight pavement edge with grass flopping over it, hiding the true edge line.

Right sketch — confirmed, at small scale. Best case a1_b000_s049 below: grass blades visibly lean 0.1–0.3 m over the rim, and the flat pavement surface demonstrably continues beneath the fringe (pavement-height LiDAR returns under the blades). Seen clearly at 1 site, grazing/marginal at 3 more — always decimetre-scale.
Left sketch — not found. 0 of 65 sites show the pavement itself indenting. 59× "no", 6× "unclear" (data too thin to prove either way), 0× confirmed. The v1 claim of 18% real notches did not survive re-examination against the cross-sections.

What produces deep "notches" instead: the detector leaves the physical edge entirely (lane markings, scan seams, intensity boundaries, occlusion holes) — see detector artifacts.

Percentages (v2, detector ignored)

66%
43 of 65
Detector artifact
15%
10 of 65
Vegetation at the edge (grass/verge)
8%
5 of 65
Object (signs, posts, guardrail)
6%
4 of 65
Unpaved shoulder transition
5%
3 of 65
Dropout / unclear
Detector artifact Vegetation at edge Object Unpaved shoulder Dropout/unclear

The vegetation 10, broken down

Sub-modenSites
Grass clearly overhanging pavement, pavement continuing beneath (the right-hand sketch)1a1_b000_s049
Grass/verge fringe grazing the rim (≤ ~0.2 m, at resolution limit)3s032, s062, s119 (a1_b000)
Hedge abutting the edge, no overhang over asphalt1a1_b000_s045
Vegetation near edge, A2/A3 — no point cloud, unverifiable5a2_s082/s110/s137, a3_s252, a1_b000_s083*

*s083: raised verge with small posts/shrub stems — vegetation present but nothing over pavement. So the direct answer to "in what percentage of the edges is vegetation overhanging?" at these notch sites: visible grass at/over the rim ≈ 15%; demonstrable overhang with pavement beneath ≈ 2–6%, always ≤ ~0.3 m. (Percentages are per sampled notch site, not per metre of edge — a per-metre sweep would need a dedicated pass.)

Real notch in the asphalt: 0 confirmed / 6 unclear / 59 no.

Grass-overhang walkthroughs (the right-hand sketch)

grass overhang — confirmed a1_b000_s049_L_090.25

The clearest case in the whole sample. Perspective close-up: green blades spill from the verge onto the pale asphalt; asphalt is visible between and around the tufts. Cross-section: a ~0.3–0.4 m tall grass clump at lateral −1.5…0 whose blades arc over the rim at 0 while the flat pavement line continues straight beneath the fringe. Canopy-stripped top-down: grey pavement-level returns run right up to (and mixed into) the grass line — the edge under the grass is straight; the detected polyline (red) bulges inboard exactly over the tuft. Overhang ≈ 0.1–0.3 m over an intact edge.

1 · Perspective close-up (~3.5 m of scene): grass blades over pale asphalt at the mark. Ignore red/cyan overlays (detector).
2 · Cross-section at notch centre (0.3 m slab): blades arc over lateral 0; flat pavement continues beneath. Left spike = post.
3 · Canopy-stripped top-down, ±2.2 m: pavement-level grey right up to the grass speckle line; red polyline bulges inboard over it.
4 · Same crop, all points: the tuft in true RGB.
5 · Context, 10 m box.
grass fringe grazing the rim a1_b000_s062_R_058.50 · s032_R_032.75 · s119_L_019.75

The recurring A1 pattern: a raised grass verge ramps up 0.15–0.3 m directly from an intact pavement edge; tufts lean onto the last decimetres of asphalt but nothing hangs clear of it. In s032 the pavement genuinely ends at the rim with a step-down to an embankment — grass sits above the rim, but there is no hidden asphalt beneath (v1's "pavement under canopy = yes" was wrong here). In all three the asphalt edge itself is straight; the notch is the detector reacting to the occluded/raised verge.

s062 perspective close-up: mark sits on dark asphalt just inboard of the grass strip; modest tufts at the boundary.
s062 cross-section: verge scatter ramps up continuously from the edge — solid ground, not canopy; thin spike = stem/post.
s032 perspective close-up: grass band between asphalt and guardrail; tufts encroach slightly onto the asphalt.
s032 cross-section: pavement ends at 0 with a small step-down to the embankment; grass cluster above the rim, nothing beneath.
s119 cross-section: same raised-verge shape; pavement continuous through the "notch".
s119 canopy-stripped top-down ±2.2 m.

Abutting, not overhanging

hedge abuts the edge a1_b000_s045_L_047.75 · and s083_R_082.00, a3_s110_R_028.50

The counterexamples that killed the "canopy overhang" reading. s045: a ~1 m hedge is massed against the edge, but its foliage stops within ~0.1–0.2 m of where the asphalt ends — it abuts the rim, it does not reach over the roadway, and there is no pavement continuing under it (verge grass, then nothing at pavement level). s083: raised rough verge with a couple of narrow ~0.5 m posts/shrub stems. a3_s110: the "vegetation" at the candidate is actually a road-sign panel on a post — an object.

s045 cross-section: dense hedge mass from −3 to ≈−0.2; pavement flat to lateral ~0; hedge edge and asphalt edge nearly coincide — abutment.
s045 perspective close-up.
s045 top-down: hedge band starts exactly at the edge; road above is intact.
s083 cross-section: raised verge + two thin posts; pavement ends normally.
a3_s110: the object at the mark is a sign panel on a post, not a bush.

Detector artifacts — the dominant mode (66%)

Typical members of the 43: the polyline detours across intact carriageway (rectangular excursions onto lane markings/scan seams), or wanders through a paved apron / gravel shoulder where no edge exists. Nothing at these locations resembles either sketch — the asphalt is simply continuous where the "notch" is drawn.

s022: detected edge draws a 6 m rectangle mid-carriageway on intact asphalt. True verge is at the bottom of frame.
s048: a 5 m "notch" drawn across continuous pavement.
s048 cross-section: uninterrupted flat asphalt for ~8 m through the claimed notch.
s017: asphalt → wide gravel/dirt apron; polylines are rectangles with no physical meaning. No vegetation at all.

Method & caveats

Data