Source Journal of Chinese Scientific and Technical Papers
Included as T2 Level in the High-Quality Science and Technology Journals in the Field of Architectural Science
Core Journal of RCCSE
Included in the CAS Content Collection
Included in the JST China
Indexed in World Journal Clout Index (WJCI) Report
Volume 55 Issue 5
May  2025
Turn off MathJax
Article Contents
WANG Wei, MA Xinyu, PENG Fan, ZENG Kai. Research on 3D Data Restoration of Industrial Heritage Based on Depth Image: A Case Study of Changsha Kaixue Flour Mill[J]. INDUSTRIAL CONSTRUCTION, 2025, 55(5): 37-45. doi: 10.3724/j.gyjzG25012802
Citation: WANG Wei, MA Xinyu, PENG Fan, ZENG Kai. Research on 3D Data Restoration of Industrial Heritage Based on Depth Image: A Case Study of Changsha Kaixue Flour Mill[J]. INDUSTRIAL CONSTRUCTION, 2025, 55(5): 37-45. doi: 10.3724/j.gyjzG25012802

Research on 3D Data Restoration of Industrial Heritage Based on Depth Image: A Case Study of Changsha Kaixue Flour Mill

doi: 10.3724/j.gyjzG25012802
  • Received Date: 2025-01-28
    Available Online: 2025-07-15
  • Due to inadequate conservation frameworks, substantial industrial heritage in China has been demolished or damaged, necessitating digital reconstruction to preserve historical information. An approach is proposed for 3D data restoration of demolished industrial heritage with incomplete UAV photogrammetry datasets. Depth image technology refines aerial imagery through depth-aware analysis to optimize point clouds, enhancing completeness in information-deficient scenarios. By integrating historical records, remote sensing, and archival photos, detailed point cloud refinement and scene reconstruction are achieved. The resulting 3D models enable morphological restoration of urban industrial architecture, providing a systematic framework for digital heritage preservation and database development using depth-image-based data recovery techniques.
  • loading
  • [1]
    徐苏斌,张晶玫,田培培. 中国近代工业遗产分类谱系溯源研究[J]. 工业建筑,2024,54(3):1-8.
    [2]
    刘抚英,宋智强,强唯,等. 杭嘉湖地区近现代丝绸工业遗产综合信息数据库构建与应用研究[J]. 工业建筑,2023,53(6):91-99.
    [3]
    郭黛姮. 圆明园研究回顾与展望[C]// 数字化视野下的圆明园. 北京:2010.
    [4]
    WEI W,HEI M,PENG F,et al. Development of“air-ground data fusion” based LiDAR method towards sustainable preservation and utilization of multiple-scaled historical blocks and buildings[J]. Sustainable Cities and Society,2023(9),104414.
    [5]
    王蔚,彭凡,李璟,等. 空地一体化历史建筑三维重建与再利用研究:以长沙潮宗街金九故居为例[J]. 城市建筑,2022,19(13):135-142.
    [6]
    赵立培,童文喜. 点云补全技术的研究与应用[J]. 现代计算机,2024,30(17):30-36.
    [7]
    秦旭元,赵志林. 无人机倾斜摄影技术的精细化建模[J]. 北京测绘,2024,38(4):509-513.
    [8]
    陈汝杰. 多阈值深度图像分割算法研究[D]. 广州:广东工业大学,2023.
    [9]
    杨宜林. 基于深度图像的三维点云配准算法研究[D]. 兰州:兰州交通大学,2021.
    [10]
    刘瑞,吕开云,袁志聪,等. 基于深度图像与分水岭的平面点云分割方法[J]. 江西科学,2021,39(1):166-171.
    [11]
    长沙凯雪为毛主席做过寿面 是全国仅有的 3 家百年面粉企业之一[EB/OL].[ 2013-02-28

    ]. http://www.changsha0731.cn/site/portal.php?mod=view&aid=7079.
    [12]
    黄凰,程淑媛. 近代工业遗产保护与再利用研究:以芜湖益新面粉厂为例[J]. 文化软实力研究,2024,9(5):91-99.
    [13]
    张志宽. 基于深度图像预估的三维点云重建算法研究[D]. 哈尔滨:哈尔滨工业大学,2020.
    [14]
    王浩,张叶,沈宏海,等. 图像增强算法综述[J]. 中国光学,2017,10(4):438-448.
    [15]
    YANG L,KANG B,HUANG Z,et al. Depth anything:unleashing the power of large-scale unlabeled data[J]. IEEE,2024. DOI:10.1109/CVPR52733.2024.00987.
    [16]
    YANG L,KANG B,HUANG Z,et al. Depth Anything V2[C]// 38th Conference on Neural Information Processing Systems(Neur IPS 2024). arXiv:2406.09414,2024.
    [17]
    郭铁柱. 深度图像的填充与点云数据的优化[D]. 北京:北京工业大学,2014.
    [18]
    YAO X,HU L,MA Y,et al. Unsupervised single image-based depth estimation powered by coplanarity-driven disparity derivation[J]. Engineering Applications of Artificial Intelligence,2024,138(Dec.Pt.B):109432.1-109432.13.
    [19]
    张宁. 基于深度图像的点云配准技术研究[D]. 太原:中北大学,2017.
    [20]
    于双. 基于点云逆向建模的历史建筑数字化复原技术研究[J]. 测绘与空间地理信息,2023,46(7):148-150.
    [21]
    张慧,汪杰. 基于点云数据的传统村落VR展示研究:以井陉于家石头村为例[J]. 古建园林技术,2023(1):48-51.
    [22]
    石越. BIM在工业遗产信息采集与管理中的应用[D]. 天津:天津大学,2014.
  • 加载中

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Article Metrics

    Article views (144) PDF downloads(5) Cited by()
    Proportional views
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return