东南大学学报（自然科学版）2025，Vol.55Issue(5)：1467-1475,9.DOI:10.3969/j.issn.1001-0505.2025.05.029
基于机器视觉和边缘重构的拉索振动响应识别
Vibration response identification of cable based on machine vision and edge reconstruction
摘要
Abstract
To realize non-contact and high-precision measurement of cable vibration response,an improved method for cable edge identification was proposed.Based on the gray characteristics of the region,the Lorentz function was used to fit the gray extremum of the edge region,and the subpixel maximum point was obtained for the construction of morphological edge of cable.Model and field tests were carried out in the laboratory and Maanshan Yangtze River highway and railway bridge respectively to verify the measurement effect of the method under various illumination conditions.The results show that the cable vibration response identification method based on edge reconstruction is adaptable under various illuminations.Under high illumination condi-tions,the maximum error of vibration response measurement in the model experiment is about 1.5%.Under natural light with low light illumination conditions,the proposed method also has good adaptability and the er-ror in the field tests is about 1.5%.The vibration response measurement method based on machine vision has the advantage of non-contact measurement,which can provide a new solution for the vibration measurement of cable.关键词
机器视觉/边缘重构/斜拉索振动/结构监测Key words
machine vision/edge reconstruction/cable vibration/structure monitoring分类
交通工程引用本文复制引用
杜文康,王浩,雷冬,李冰,倪有豪..基于机器视觉和边缘重构的拉索振动响应识别[J].东南大学学报（自然科学版）,2025,55(5):1467-1475,9.基金项目
国家自然科学基金资助项目(52338011). （52338011）
