Research Article
Damage Identification Method of Building Structure Based on Computer Vision
@INPROCEEDINGS{10.1007/978-3-031-50574-4_13, author={Hongyue Zhang and Xiaolu Deng and Guoliang Zhang and Xiuyi Wang and Longshuai Liu and Hongbing Wang}, title={Damage Identification Method of Building Structure Based on Computer Vision}, proceedings={Multimedia Technology and Enhanced Learning. 5th EAI International Conference, ICMTEL 2023, Leicester, UK, April 28-29, 2023, Proceedings, Part II}, proceedings_a={ICMTEL PART 2}, year={2024}, month={2}, keywords={Computer Vision Building Structure Structural Damage Damage Identification}, doi={10.1007/978-3-031-50574-4_13} }- Hongyue Zhang
Xiaolu Deng
Guoliang Zhang
Xiuyi Wang
Longshuai Liu
Hongbing Wang
Year: 2024
Damage Identification Method of Building Structure Based on Computer Vision
ICMTEL PART 2
Springer
DOI: 10.1007/978-3-031-50574-4_13
Abstract
Under the influence of load, earthquake, settlement and other factors, the building structure will be damaged to different degrees. If the building structure damage is not found and handled in time, it will lead to the risk of building components falling off or even collapsing. Therefore, a method of building structure damage identification based on computer vision is proposed. According to different damage types and mechanisms of building structures, the identification criteria of building structure damage are set up to provide reference for damage identification. The computer vision technology is used to collect the building structure image, and through geometric registration, light correction, graying, noise reduction and other steps, complete the preprocessing of the initial building structure image, improve the image denoising effect, and avoid the impact of noise, light and other factors on the recognition results. The image features of the building structure are extracted from the three aspects of color, texture and geometric shape, and compared with the set recognition standards. The damage type of the building structure is determined by similarity measurement, and the identification results including the damage parameters of the building structure are output. Compared with the traditional identification methods, the identification accuracy of the optimized design building damage identification method is improved, and the parameter identification error is smaller, that is, the identification performance is better.
