← back to the atlas

Seafloor image → nodule coverage: detector v3

A convolutional network (ResNet-34, 1120-px images) trained to read nodule coverage (%) directly from seafloor photographs, using the GEOMAR SO268 OFOS corpus (Clarion-Clipperton Zone, 4,000–4,500 m; Purser et al. 2021 / Mbani & Greinert 2023, CC-BY) with per-image coverage labels from the CoMoNoD algorithm (Schoening 2021). Trained on the full corpus on a cloud GPU: 19,683 training images, tested on 9,096 images from four whole profiles the model never saw.

Held-out accuracy

ModelMAE (coverage %)R²
CNN v3, best epoch (8)2.95%0.65
Classical dark-blob thresholding (calibrated)3.4%0.53
Predict the training mean10.3%0
predicted vs labelled coverage

By held-out profile

ProfileAreaMean coverageImagesMAER²
SO268/2_100-1German13%2,4951.35%0.71
SO268/2_128-1Belgian30%1,9348.05%0.28
SO268/2_147-1Belgian18%2,4942.13%0.88
SO268/2_177-1German13%2,1731.18%0.82

On sparse and medium fields the model reads coverage within one to two percentage points, image by image. The one packed profile (30% mean coverage) is the open problem: the class mean is right but per-image error is 8%. Part of that is label noise — the labelling algorithm itself is least reliable in packed facies — and it is the same failure mode TMC reported for its 2018 tooling.

Along-track view (earlier v1 model, four held-out profiles)

transects

Training curve (v3)

EpochHoldout MAER²
15.74%0.43
25.57%0.25
34.23%0.58
44.49%0.59
53.62%0.31
62.96%0.63
73.91%0.35
82.95%0.65
93.00%0.61
102.99%0.59
113.16%0.60
122.99%0.61

Why it matters

Coverage from imagery is the dense, cheap signal; box cores are the sparse, expensive truth. A detector like this turns every AUV or towed-camera transect into a continuous coverage map, which is the input to a CRIRSCO/ISA Annex V-style abundance block model once co-located core photos calibrate coverage to kg/m². Our first calibration test on the 36 SO268 box cores showed that track-average coverage alone does not predict core abundance — the kg/m² step needs per-core scaled photos and burial covariates.

Draft, 2026-09-21 — Bryce Judy. Images and labels © their authors under CC-BY 4.0; model outputs are ours. Atlas.