Geometric recognition methodology of honeycomb structure based on dynamic window

Abstract

Defect determination is important for guaranteeing the quality of sandwich structure. The present study, a geometric recognition methodology based on dynamic window was constructed for honeycomb structure, by identifying the vertices from given images captured in the manufacturing process. The combination of traversing window, annihilation window, and tracking window was developed to reconstruct the contoured maximums, locate the vertices positions, and determine the cell–node relationships, respectively. Moore boundary tracking was practically used to deal with the problem caused by the edge effect. Based on these, all relevant vertices as well as their relationship of each honeycomb cell can be determined. Afterward, quality assessment of the honeycomb product was carried out in terms of the maximum, the average, and the maximum average degree deviation. Illustrations validate the present methodology well. All these achievements shed a light on design of the consistency, reliability, and homogeneity of high-standard sandwich structure.