SafeRestore: Detector-Relative Risk Certificates for Selective Industrial Image Restoration
Abstract
Industrial inspection pipelines often restore a measured image before a detector acts on it, yet restoration can suppress detector-supported defect structure or create clean-region activations. We formulate restoration as a selective action problem over the measured display, five restored candidates, and review. SafeRestore ranks candidates with action-specific fitted scores, chooses a gate on threshold-tuning data, and evaluates the fixed gate on a disjoint certification sample with two one-sided exact binomial bounds: one for the positive-conditional evidence-loss incident rate and one for the all-accepted excess-activation incident rate. The guarantee is marginal for one policy fixed before its certification outcomes are observed, under an image-level i.i.d. working model. In a retrospective split-sample study of 4,591 public Carinthia-S images, the protocol yields auditable risk-coverage behavior. The primary all-action policy passes in one of five training repetitions (12.0% +/- 26.9% pass-gated test coverage when failures count as zero), whereas fixed bicubic and reduced-complexity variants pass more often. On reserved morphologies, evidence-loss incidence rises to 81.1-90.3%, and KolektorSDD lacks both detector competence and enough positive certification images for the stated target. The contribution is therefore an auditable, detector-relative framework for deciding when a transformed image may be returned automatically and when review remains necessary -- not a claim that adaptive routing outperforms simpler policies on the present evidence.
Related
- How Many Labels Does Model Choice Need? Certificates and Budgets for Selective Prediction
- Optimizing Explicit Unit-Distance Lower-Bound Certificates
- Unique Normal Forms: Proof-Graph Counterexamples, Infinite-Carrier Certificates, and Confluence Repair
- Sum-of-Squares Certificates for Copositive Matrices via Recursive Identities: The de Klerk-Pasechnik Conjecture and Hoffman-Pereira Matrices
- Task-aware Kernel Stein certificates for Rényi differential privacy
BibTeX
@misc{yang2026saferestore,
title = {{SafeRestore: Detector-Relative Risk Certificates for Selective Industrial Image Restoration}},
author = {Shaoliang Yang and Jun Wang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.03475},
eprint = {2609.03475},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.03475},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - SafeRestore: Detector-Relative Risk Certificates for Selective Industrial Image Restoration
AU - Shaoliang Yang
AU - Jun Wang
PY - 2026
DA - 2026/09/03
PB - arXiv
DO - 10.48550/arxiv.2609.03475
UR - https://arxiv.org/abs/2609.03475
AB - Industrial inspection pipelines often restore a measured image before a detector acts on it, yet restoration can suppress detector-supported defect structure or create clean-region activations. We formulate restoration as a selective action problem over the measured display, five restored candidates, and review. SafeRestore ranks candidates with action-specific fitted scores, chooses a gate on threshold-tuning data, and evaluates the fixed gate on a disjoint certification sample with two one-sided exact binomial bounds: one for the positive-conditional evidence-loss incident rate and one for the all-accepted excess-activation incident rate. The guarantee is marginal for one policy fixed before its certification outcomes are observed, under an image-level i.i.d. working model. In a retrospective split-sample study of 4,591 public Carinthia-S images, the protocol yields auditable risk-coverage behavior. The primary all-action policy passes in one of five training repetitions (12.0% +/- 26.9% pass-gated test coverage when failures count as zero), whereas fixed bicubic and reduced-complexity variants pass more often. On reserved morphologies, evidence-loss incidence rises to 81.1-90.3%, and KolektorSDD lacks both detector competence and enough positive certification images for the stated target. The contribution is therefore an auditable, detector-relative framework for deciding when a transformed image may be returned automatically and when review remains necessary -- not a claim that adaptive routing outperforms simpler policies on the present evidence.
ER -