杨娜, 张翀, 李天昊. 基于无人机与计算机视觉的中国古建筑木结构裂缝监测系统设计[J]. 工程力学, 2021, 38(3): 27-39. DOI: 10.6052/j.issn.1000-4750.2020.04.0263
引用本文: 杨娜, 张翀, 李天昊. 基于无人机与计算机视觉的中国古建筑木结构裂缝监测系统设计[J]. 工程力学, 2021, 38(3): 27-39. DOI: 10.6052/j.issn.1000-4750.2020.04.0263
YANG Na, ZHANG Chong, LI Tian-hao. DESIGN OF CRACK MONITORING SYSTEM FOR CHINESE ANCIENT WOODEN BUILDINGS BASED ON UAV AND CV[J]. Engineering Mechanics, 2021, 38(3): 27-39. DOI: 10.6052/j.issn.1000-4750.2020.04.0263
Citation: YANG Na, ZHANG Chong, LI Tian-hao. DESIGN OF CRACK MONITORING SYSTEM FOR CHINESE ANCIENT WOODEN BUILDINGS BASED ON UAV AND CV[J]. Engineering Mechanics, 2021, 38(3): 27-39. DOI: 10.6052/j.issn.1000-4750.2020.04.0263

基于无人机与计算机视觉的中国古建筑木结构裂缝监测系统设计

DESIGN OF CRACK MONITORING SYSTEM FOR CHINESE ANCIENT WOODEN BUILDINGS BASED ON UAV AND CV

  • 摘要: 中国古建筑木结构中裂缝繁多,裂缝成因与发展规律复杂,易引起木结构构件脆断,从而严重威胁中国古建筑木结构健康情况。该文基于无人机与计算机视觉技术设计了一套适用于中国古建筑木结构裂缝的监测系统,该监测系统包含无人机系统、相机系统和图像处理系统。在无人机系统中,该文设计了一款适合于中国古建筑木结构裂缝监测的无人机,并分析其悬停拍照的可行性。在相机系统中,进行了相机畸变矫正、像素解析度标定,并提出了一种改进的SIFT+RANSAC方法以提高裂缝图像拼接精度。在图像处理系统中,选择了适用于中国古建筑木结构裂缝图像的预处理方式,并将Hessian矩阵与自适应阈值分割法融合,有效地提取了中国古建筑木结构裂缝特征,进而通过计算机视觉测量方法准确识别构件和裂缝的尺寸。最后,基于中国古建筑木结构亭子模型验证了所提出中国古建筑木结构裂缝监测系统的可行性。

     

    Abstract: There are many cracks in Chinese ancient wooden buildings, and the formation and development of these cracks are complicated. These cracks can easily cause component brittleness, which seriously threaten the condition of Chinese ancient wooden buildings. Based on an unmanned aerial vehicle (UAV) and computer vision (CV), it designs a crack monitoring systems for Chinese ancient wooden structures, including an UAV system, a camera system, and an image processing system. In the UAV system, a UAV suitable for crack monitoring is designed, and the feasibility of hovering photography of the UAV is analyzed. In the camera system, the camera distortion correction and pixel resolution calibration are done, and an improved SIFT + RANSAC method is proposed to improve the accuracy of crack image mosaic. In the image processing system, a set of preprocessing methods is adopted, and Hessian matrix is combined with adaptive threshold segmentation algorithm, extracting crack features effectively. Then the size of members and cracks are accurately identified by CV measurement method. Finally, the feasibility of the proposed crack monitoring system is verified based on the wooden pavilion model of Chinese ancient buildings.

     

/

返回文章
返回