Deng entropy and LSTM neural network to classify built cultural heritage with severe damage
Source abstract
Abstract
• An automatic approach for 3D recognition of structural state of uniquely built cultural heritage with severe and very severe damage is proposed. • Deng entropy and information dimension computed from voxel maps of the 3D reconstruction of built cultural heritage are exploited. • Deng entropy sequences are processed by bidirectional Long-Short Term Memory neural network yielding successful training and test results. • The approach is successfully applied to recognize very severe damage of a unique Mexican cultural heritage. This paper proposes an automatic approach for the 3D recognition of the structural state of uniquely built cultural heritage with severe and very severe damage degree. Low resolution voxel map of 3D reconstruction (point cloud), Deng entropy, and bidirectional LSTM neural network are exploited. A practical test confirms the effectiveness of the proposed approach. This paper proposes an automatic approach for the 3D recognition of the structural state of uniquely built cultural heritage with severe and very severe damage. For this purpose, a 3D reconstruction (point cloud) mapped to voxels of built cultural heritage is used. Then, Deng entropy is computed for the resulting voxel map, and the corresponding information dimension is obtained by applying the box-covering method. The entropy and information dimension data sequences are combined into one sequence, called the Deng sequence. This is in order to quantify the occupied space in voxel map of built cultural heritage. Deng sequence is used as the input for a bidirectional Long Short-Term Memory neural network (bLSTMnn) to automatically classify the structural state of built cultural heritage. The main consideration is that few models of built cultural heritage are available due to their uniqueness. Hence, a small dataset created by authors using Computer-Aided Design tools is used for the training of the bLSTMnn. A cross-method is applied to validate the bLSTMnn. The proposed approach is successfully applied to recognize the current structural state of a unique Mexican cultural heritage building with very severe damage.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Infrastructure Maintenance and Monitoring · Conservation Techniques and Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (39)
Show source-provided references
1. World heritage sites and museums: a pact for sustainable development, https://whc.unesco.org/en/activities/941/#:~:text=World%20Heritage%20sites%20are%20vital,for%20the%20generation%s%20to%20come. Accessed 10-feb-2025.
2. Artificial intelligence-based automatic visual inspection system for built heritage10.1108/SASBE-09-2020-0139 ↗
3. Smart monitoring solution through internet of things utilization to achieve resilient preservation
4. CHPC: a complex semantic-based secured approach to heritage preservation and secure IoT-based museum processes10.1016/j.comcom.2019.08.001 ↗
5. Three-dimensional reconstruction of national traditional sports cultural heritage based on feature clustering and artificial intelligence
6. Resolving deterioration of heritage building elements using an expert system10.1108/IJBPA-12-2019-0106 ↗
7. ART-RISK 3.0 a fuzzy-based platform that combine GIS and expert assessments for conservation strategies in cultural heritage10.1016/j.culher.2022.03.012 ↗
8. Deep learning for 3d point clouds: a survey10.1109/TPAMI.2020.3005434 ↗
9. Deep learning in food category recognition10.1016/j.inffus.2023.101859 ↗
10. Object detection in 20 years: a survey10.1109/JPROC.2023.3238524 ↗
11. Detection of disaster-affected cultural heritage sites from social media images using deep learning techniques10.1145/3383314 ↗
12. Deep learning in historical architecture remote sensing: automated historical courtyard house recognition in Yazd, Iran10.3390/heritage5040159 ↗
13. Feature recognition of regional architecture forms based on machine learning: a case study of architecture heritage in Hubei Province, China10.3390/su15043504 ↗
14. Saliency-based selection of visual content for deep convolutional neural networks: application to architectural style classification10.1007/s11042-018-6515-2 ↗
15. Architectural style classification of mexican historical buildings using deep convolutional neural networks and sparse features10.1117/1.JEI.26.1.011016 ↗
16. Explainable neural-symbolic learning (X-NeSyL) methodology to fuse deep learning representations with expert knowledge graphs: the monumai cultural heritage use case10.1016/j.inffus.2021.09.022 ↗
17. Research on image classification and retrieval using deep learning with attention mechanism on diaspora chinese architectural heritage in Jiangmen, China10.3390/buildings13020275 ↗
18. Transfer learning and performance enhancement techniques for deep semantic segmentation of built heritage point clouds10.4995/var.2021.15318 ↗
19. An artificial neural network framework for classifying the style of cypriot hybrid examples of built heritage in 3D10.1016/j.culher.2023.07.016 ↗
20. A 3D digitisation workflow for architecture-specific annotation of built heritage
21. 3DLEB-Net: Label-efficient deep learning-based semantic segmentation of building point clouds at LoD3 level10.3390/app11198996 ↗
22. Implementing pointnet for point cloud segmentation in the heritage context10.1186/s40494-022-00844-w ↗
23. Classification of architectural heritage images using deep learning techniques10.3390/app7100992 ↗
24. Machine learning models for cultural heritage image classification: comparison based on attribute selection10.3390/info11010012 ↗
25. Artificial intelligence-based visual inspection system for structural health monitoring of cultural heritage
26. Crack detection in historical structures based on convolutional neural network
27. Deca convolutional layer neural network (DCL-NN) method for categorizing concrete cracks in heritage building
28. Damage classification for masonry historic structures using convolutional neural networks based on still images10.1111/mice.12411 ↗
29. Deep learning-based weathering type recognition in historical stone monuments10.1016/j.culher.2020.04.008 ↗
30. Autonomous damage segmentation and measurement of glazed tiles in historic buildings via deep learning10.1111/mice.12488 ↗
31. Surface damage identification for heritage site protection: a mobile crowd-sensing solution based on deep learning10.1145/3569093 ↗
32. CNN-based statistics and location estimation of missing components in routine inspection of historic buildings10.1016/j.culher.2019.02.002 ↗
33. Application of deep learning approach for the classification of buildings’ degradation state in a BIM methodology10.3390/app12157403 ↗
34. Behaviour of foundations and structures
35. Museo y centro de documentación, ex convento de la natividad en tepoztlán, https://mediateca.inah.gob.mx/islandora_74/islandora/object/museo:1520. Accessed 21-mar-2024.
36. Primeros monasterios del siglo xvi en las laderas del popocatépetl, https://whc.unesco.org/es/list/702. Accessed 21-mar-2024.
37. Deng entropy10.1016/j.chaos.2016.07.014 ↗
38. Deng entropy and information dimension for COVID-19 and common pneumonia classification10.1142/S0218348X24500336 ↗
39. A. Ramirez-Arellano, P. Ortiz-Vilchis, J. Bory-Reyes, The role of D-summable information dimension in differentiating COVID-19 disease, Fractals 29 (8) (2021) 2150255, doi:10.1142/S0218348X21502558.10.1142/S0218348X21502558 ↗