Abstract. As there usually exist widespread crack, decay, deformation and other damages in the wooden architectural heritage (WAH). It is of great significance to detect the damages automatically and rapidly in order to grasp the status for daily repairs. Traditional methods use artificial feature-driven point clouds and image processing technology for object detection. With the development of big data and GPU computing performance, data-driven deep learning technology has been widely used for monitoring WAH. Deep learning technology is more accurate, faster, and more robust than traditional methods.In this paper, we conducted a case study to detect timber-crack damages in WAH, and selected the YOLOv3 algorithm with DarkNet-53 as the backbone network in the deep learning technology according to the characteristics of the crack. A large timber-crack dataset was first constructed, based on which the timber-crack detection model was trained and tested. The results were analyzed both qualitatively and quantitatively, showing that our proposed method was able to reach an accuracy of more than 90% through processing each image for less than 0.1s. The promising results illustrate the validity of our self-constructed dataset as well as the reliability of YOLOv3 algorithm for the crack detection of wooden heritage.
<p><strong>Abstract.</strong> A demand-oriented Building Information Model (BIM) model built using high-fidelity point cloud data can better protect architectural heritage. The multi-level detail (mutli-LoD) parametric model emphasizes the different protection requirements of typical components and the automatic extraction of corresponding parameters of high-fidelity point clouds, which are two related key issues. Taking the typical Chinese wooden architectural heritage as an example, according to different requirements, the multi-LoD principle of typical components is proposed. On this basis, the automatic extraction method of the above parameters is developed, and the key parameters of the method are recommended. In order to solve the above problems, taking the three typical Dou-Gong used in Liao Dynasty and Song Dynasty, including Zhutou Puzuo, Bujian Puzuo and Zhuanjiao Puzuo, as an example, briefly introduced the standardization characteristics of the typical components of the "Yingzao Fashi". Subsequently, the corresponding multiple LoD principles are recommended according to different requirements. Based on this and high-fidelity point cloud data, an automatic extraction method for multi-LoD BIM model parameters for typical components of wooden architectural heritage is proposed.</p>
Abstract. Chinese ancient architecture is a valuable heritage wealth, especially for roof that reflects the construction age, structural features and cultural connotation. Point cloud data, as a flexible representation with characteristics of fast, precise, non-contact, plays a crucial role in a variety of applications for ancient architectural heritage, such as 3D fine reconstruction, HBIM, disaster monitoring etc. However, there are still many limitations in data editing tasks that need to be worked out manually, which is time-consuming, labor-intensive and error-prone. In recent years, the theoretical advance on deep learning has stimulated the development of various domains, and digital heritage is not in exception. Whenever, deep learning algorithm need to consume a huge amount of labeled date to achieve the purpose for segmentation, resulting a actuality that high labor costs also be acquired. In this paper, inspired by the architectural style similarity between mimetic model and real building, we proposed a method supported by deep learning, which aims to give a solution for the point cloud automatic extraction of roof structure. Firstly, to generate real point cloud, Baoguang Temple, unmanned Aerial Vehicle (UAV) is presented to obtain image collections that are subsequently processed by reconstruction technology. Secondly, a modified Dynamic Graph Convolutional Neural Network (DGCNN) which can learn local features with taking advantage of an edge attention convolution is trained using simulated data and additional attributes of geometric attributes. The mimetic data is sampled from 3DMAX model surface. Finally, we try to extract roof structure of ancient building from real point clouds scenes utilizing the trained model. The experimental results show that the proposed method can extract the rooftop structure from real scene of Baoguang, which illustrates not only effectiveness of approach but also a fact that the simulated source perform potential value when real point cloud datasets are scarce.
No abstract
No abstract
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.