Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper presents a novel framework for sand-boil segmentation in earthen levees, addressing the common pitfalls of inflated accuracy due to improper training and validation practices. By employing a leak-free cross-validation strategy and a per-architecture calibration approach, the authors ensure that synthetic images used for training are appropriately linked to their real counterparts, thus maintaining the integrity of the evaluation process. The Updated SandBoilNet achieves a significant improvement in intersection-over-union metrics, reaching 0.707 on the held-out test set, outperforming previous methods while also introducing a mask-conditioned synthesis technique for generating labeled training images at no annotation cost.
Achieving a 0.707 intersection-over-union in sand-boil segmentation, this framework effectively eliminates common biases in model training and evaluation.
Sand boils, points where water seeping beneath an earthen levee re-emerges at the surface, are early warnings of internal erosion, and deep segmentation networks are increasingly used to find them in inspection photographs. Annotated examples are scarce, and two common ways of working around that scarcity quietly inflate reported accuracy: tuning ensemble weights on the same images later used to score them, and training on synthetic images derived from the very photographs held out for testing. We present a sand-boil segmentation framework that closes both loopholes. Every synthetic image carries a pointer to its real parent, and a per-fold filter excludes any image whose parent is held out; five encoder-decoder backbones are trained under five-fold cross-validation, calibrated by one temperature scalar each, and combined by a per-pixel meta-learner fitted only on out-of-fold predictions. On the held-out test set the proposed Updated SandBoilNet reaches an intersection-over-union of 0.707 over three seeds, against 0.608 for the published original re-evaluated on the same split. Under the stacking protocol the calibrated stack reaches 0.681 against 0.694 for the strongest fold-averaged member, so it does not improve on the best single model; eight meta-learner families reproduce that outcome, which we trace to a mean pairwise error correlation of 0.894 among members. A synthetic pool filtered for label fidelity lifts the champion to 0.718 over three seeds against a 0.707 control. We also introduce a mask-conditioned synthesis route that makes the conditioning mask the label by construction, giving labelled training images at zero annotation cost.