A Multiscale Structured-Light Camera System for Quantifying Machining Quality of Honeycomb Composites

Abstract

Honeycomb composites are widely used as lightweight load-bearing components in aerospace applications. However, their thin-walled hexagonal cell structures are prone to defects, such as burrs, tearing, crushing, and deformation during machining. These will compromise structural integrity and thus require precise measurement. Due to the multiscale distribution of these defects, detection remains challenging: microscale burrs are often too fine to capture, mesoscale damage lacks clear patterns, and macroscale surfaces are large and curved, limiting efficient evaluation. The existing methods, such as optical microscopy and line laser measurement, are limited to single-scale detection. Switching between different equipment is time-consuming and introduces inconsistencies. To address these issues, this study proposes a unified, multiscale detection platform based on a structured light camera (SLC). A multiscale classification and evaluation framework is developed, using customized algorithms for automated feature extraction from complex honeycomb geometries. Experiments on Nomex honeycomb composites (NHCs) demonstrate the effectiveness of the proposed method. At the microscale, burrs are quantitatively characterized using fractal dimension (FD), with the index increasing from 1.21 to 1.29 as burr complexity grows. At the mesoscale, automated defect recognition achieves accuracies of 96.6 % for tearing, 98.9 % for crushing, and 96.1 % for cell deformation. At the macroscale, the system achieves a scanning efficiency of 6750 mm2/min with a 97.2 % surface fitting accuracy even on curved structures. The methodology provides a scalable and efficient solution for quality detection and can be extended to other discontinuous composite materials.