Contour-Native Bridge Defect Detection and Compact Digital Archiving with Frequency-Supervised Fourier Contours

2026-05-09

AI Summary

This work proposes Frequency-Supervised Fourier Shape Detection (FS-FSD), a novel approach to bridge defect detection that overcomes the limitations of traditional methods—whose bounding boxes or raster masks are geometrically coarse, storage-intensive, and difficult to reuse—by introducing Fourier contour descriptors for the first time in this domain. FS-FSD leverages frequency-supervised Fourier series regression to directly output compact, reconstructible, and shareable vectorized defect boundaries. Evaluated within a unified polygonal representation space and integrated into a drone-based image processing pipeline, the method demonstrates superior performance on a large-scale dataset comprising 3,767 images and 42,346 defects, significantly outperforming existing detection, segmentation, and contour-based approaches in both polygonal accuracy and geometric fidelity.

Abstract

AI-assisted bridge defect inspection often produces bounding boxes with crude geometry or raster masks that are costly to store, transmit, and reuse. This study investigates how detected defects can be represented as compact, recoverable contour-level vector records in image space. We propose Frequency-Supervised Fourier Series Detection (FS-FSD), which directly regresses Fourier contour descriptors and evaluates boxes, masks, and contours under a unified polygon-space protocol. On 3,767 UAV-collected bridge images with 42,346 defect instances, FS-FSD achieves higher polygon-space accuracy and better matched-TP geometric quality than representative detection, segmentation, and contour baselines. These results show that, compared with bounding boxes and raster masks, Fourier contour records preserve defect-boundary geometry in a more compact, recoverable, and shareable form for engineering review and downstream information workflows. Future work will study the modeling of multi-region, fragmented, and adjacent bridge-defect boundaries and extend the framework toward long-term bridge-defect tracking and lifecycle-oriented management.

Problem

• bridge defect detection
• contour representation
• compact archiving
• Fourier contours
• polygon-space accuracy

Innovation

• Fourier contour
• bridge defect detection
• vector representation
• frequency-supervised learning
• compact digital archiving