Moonatlas

Online

Volcanic / IMP

IMP CANDIDATE

25.660°S 27.621°W · NEARSIDE · SOUTHERN HEMISPHERE

Predicted area
7,808 m²
Tile share
11.9 %
Mean softmax
0.90

uncalibrated

IMP candidateHeld-out testNASA-IBM LFM · IMP segmentation

The NASA-IBM LFM segmentation model predicts an irregular mare patch candidate covering 7,808 m² of this 256 m NAC tile. It is a model segmentation of a volcanic feature candidate, not an active volcano.

Why MOONATLAS selected this

IMP candidate

Largest predicted IMP area of its observation group. Product curation over model outputs, not a scientific ranking.

Dataset split

Held out of training and tuning, so this sample is the kind used to evaluate the model. That is evaluation provenance, not physical importance.

LROC NAC tile M1230913552LE
AI prediction · NASA-IBM LFM1.0×

M1230913552LE · 256 m tile · 1.00 m/px · masks at native resolution

A segmentation of a volcanic feature candidate, never an active volcano. The mean softmax score is uncalibrated, so it ranks pixels rather than stating a probability.

Model outputs

PREDICTED AREAAI prediction · NASA-IBM LFM
7,808 m²
IoU VS REFERENCEMOONATLAS derived metric
0.520

Provenance

Model
NASA-IBM Lunar Foundation Model
Task
Irregular mare patch segmentation on LROC NAC imagery
Checkpoint
ni_lfm_ps8_frozen_s44.ckpt
Dataset
sombench-imp-segmentation
Split
Held-out test
Source
M1230913552LE.ech.cog__target_2020__idx_2020_p0_img.tif
Processing
science-0.3.1
Data origin
Real model output
Scientific details
Model id
lfm-imp-segmentation
Model revision
10a09ffae35e
Reference
hargitai-2025
Object id
imp-m1230913552le-t2020-i2020
Catalogued
2026-09-17

Values are model output as produced, or MOONATLAS derived metrics computed from it; reference annotations are shown separately and never mixed into predictions. How MOONATLAS works.