NL-MambaXCT: Self-Supervised Nested-Learning Mamba for Nomex Honeycomb X-ray CT Defect Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Aldoboni, Ghaleb, Nassar, Lobna, Karray, Fakhri, Alshamsi, Reem
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916050995511296
author Aldoboni, Ghaleb
Nassar, Lobna
Karray, Fakhri
Alshamsi, Reem
author_facet Aldoboni, Ghaleb
Nassar, Lobna
Karray, Fakhri
Alshamsi, Reem
contents X-ray computed tomography (XCT) is widely used for non-destructive testing of Nomex honeycomb structures in aerospace manufacturing, but industrial inspection still relies heavily on manual interpretation and supervised models trained on limited labeled data. This work introduces NL-MambaXCT, a Mamba-based framework that combines self-supervised masked image modelling with a Nested Learning (NL) formulation for automated, label-efficient defect classification from production XCT slices. The backbone is a four-stage 2D encoder with RegNet convolutional blocks in the early stages and Mamba-based sequence mixing with attention in the deeper stages. It is pretrained by masked image modelling on 19,961 unlabeled industrial XCT slices and fine-tuned on 2,000 relabeled Nomex XCT slices split by production order. NL is instantiated through two-timescale parameter dynamics: selected projections maintain slow exponential-moving-average traces alongside fast weights, while a deep-momentum optimizer introduces an additional slow parameter-update trajectory. On the held-out test set, the MIM-pretrained NL-MambaXCT model achieves 96.91% accuracy and 96.8% macro F1, outperforming CNN, attention, and single-timescale Mamba baselines by 3.11--10.31 percentage points in accuracy. The results suggest that combining masked self-supervision with NL-style fast/ slow learning dynamics is a promising strategy for robust defect classification in Nomex honeycomb XCT inspection.
format Preprint
id arxiv_https___arxiv_org_abs_2605_27454
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle NL-MambaXCT: Self-Supervised Nested-Learning Mamba for Nomex Honeycomb X-ray CT Defect Classification
Aldoboni, Ghaleb
Nassar, Lobna
Karray, Fakhri
Alshamsi, Reem
Image and Video Processing
Computer Vision and Pattern Recognition
X-ray computed tomography (XCT) is widely used for non-destructive testing of Nomex honeycomb structures in aerospace manufacturing, but industrial inspection still relies heavily on manual interpretation and supervised models trained on limited labeled data. This work introduces NL-MambaXCT, a Mamba-based framework that combines self-supervised masked image modelling with a Nested Learning (NL) formulation for automated, label-efficient defect classification from production XCT slices. The backbone is a four-stage 2D encoder with RegNet convolutional blocks in the early stages and Mamba-based sequence mixing with attention in the deeper stages. It is pretrained by masked image modelling on 19,961 unlabeled industrial XCT slices and fine-tuned on 2,000 relabeled Nomex XCT slices split by production order. NL is instantiated through two-timescale parameter dynamics: selected projections maintain slow exponential-moving-average traces alongside fast weights, while a deep-momentum optimizer introduces an additional slow parameter-update trajectory. On the held-out test set, the MIM-pretrained NL-MambaXCT model achieves 96.91% accuracy and 96.8% macro F1, outperforming CNN, attention, and single-timescale Mamba baselines by 3.11--10.31 percentage points in accuracy. The results suggest that combining masked self-supervision with NL-style fast/ slow learning dynamics is a promising strategy for robust defect classification in Nomex honeycomb XCT inspection.
title NL-MambaXCT: Self-Supervised Nested-Learning Mamba for Nomex Honeycomb X-ray CT Defect Classification
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.27454