A volumetric change detection framework using UAV oblique photogrammetry – a case study of ultra-high-resolution monitoring of progressive building collapse

Xu, Ningli; Huang, Debao; Song, Shuang; Ling, Xiao; Strasbaugh, Chris; Yilmaz, Alper; Sezen, Halil; Qin, Rongjun

doi:10.1080/17538947.2021.1966527

Cited by 12 publications

(6 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The unfolded point cloud was similarly fitted to the optimal plane using the least squares method. Subsequently, the difference between each point on the point cloud surface and the optimal plane is calculated, are shown in Equation (20),…”

Section: Evaluation Of Deformationmentioning

confidence: 99%

“…The speedy development of 3D acquisition equipment has made it possible to collect 3D point cloud data of target objects efficiently, swiftly, and at high density. Relevant technologies can be mainly divided into photogrammetry, [20] light detection and ranging (LiDAR), [21] and synthetic aperture radar (SAR). [22] Photogrammetry technology generates 3D points based on stereo matching of image pairs, [23,24] which are derived through structure from motion (SfM) or multi-view stereo (MVS).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3D Change Detection Method for Exterior Wall of LNG Storage Tank Supported by Multi‐Source Spatial Data

Gao,

Guo,

Wang

et al. 2024

Advcd Theory and Sims

View full text Add to dashboard Cite

Change detection methods for measuring deformation of the LNG storage tank's exterior wall cannot meet the requirements for extensive and comprehensive deformation analysis. The method proposed in this paper uses multi‐source spatial data to detect the deformation of the LNG storage tank's exterior wall. Exterior wall data collected simultaneously by terrestrial and airborne laser scanners, as well as close‐range and oblique photogrammetry, are pre‐processed separately and then registered to the global point clouds. The deformation state of exterior wall is represented by a thorough assessment approach that takes into account four change indices. The results suggest that the registration precision of global point clouds is 6 mm. The global ovality of the LNG storage tank is 0.053%, with a tilt change of 2.58° and a deviation of 27 mm in the direction of the azimuth of 321.98°, an average value of 0.07 mm for depression and protrusion deformation, and a surface flatness of 21.3 mm. In this case, the flatness is greater than the variation threshold of the standard specification, indicating that the tank may have structural uneven settlement or other deformation problems. The attraction of the proposed method resides in its capacity to efficiently and accurately obtain spatial positioning and color information of the LNG storage tank's exterior wall, thereby providing a robust foundation for precise judgments regarding the deformation state.

show abstract

Section: Evaluation Of Deformationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

3D Change Detection Method for Exterior Wall of LNG Storage Tank Supported by Multi‐Source Spatial Data

Gao,

Guo,

Wang

et al. 2024

Advcd Theory and Sims

View full text Add to dashboard Cite

show abstract

“…Pair-wise 3D registration is the process of estimating the transformation (rigid-body transformation in our case) between a pair of 3D data sets to minimize the systematic error between two 3D data, which is widely used in 3D mapping and change detection (Xu et al, 2021) tasks. It can be categorized into coarse and fine registration according to the registration accuracy.…”

Section: Related Workmentioning

confidence: 99%

Large-scale DSM registration via motion averaging

Xu,

Qin

2024

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Generating wide-area digital surface models (DSMs) requires registering a large number of individual, and partially overlapped DSMs. This presents a challenging problem for a typical registration algorithm, since when a large number of observations from these multiple DSMs are considered, it may easily cause memory overflow. Sequential registration algorithms, although can significantly reduce the computation, are especially vulnerable for small overlapped pairs, leading to a large error accumulation. In this work, we propose a novel solution that builds the DSM registration task as a motion averaging problem: pair-wise DSMs are registered to build a scene graph, with edges representing relative poses between DSMs. Specifically, based on the grid structure of the large DSM, the pair-wise registration is performed using a novel nearest neighbor search method. We show that the scene graph can be optimized via an extremely fast motion average algorithm with O(N) complexity (N refers to the number of images). Evaluation of high-resolution satellite-derived DSM demonstrates significant improvement in computation and accuracy.

show abstract

“…Damage progression was captured from different angles at different scales during the removal of the reinforced concrete building. For damage detection from images collected during this field experiment, 44 we used a total of 523 images from cell phones and drones as the dataset for testing and evaluating the performance of APANet Mask R-CNN, since these images capture different damage levels (cracking, spalling, and mostly complete loss of concrete pieces) at various scales and different definitions. These images include three resolutions such as 1920×1080, 4030×3020 and 5470× 3640.…”

Section: An Application Of Damage Detection In a Building During A Co...mentioning

confidence: 99%

Engineering deep learning methods on automatic detection of damage in infrastructure due to extreme events

Bai

Zha

Sezen

et al. 2022

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

This paper presents a few comprehensive experimental studies for automated Structural Damage Detection (SDD) in extreme events using deep learning methods for processing 2D images. In the first study, a 152-layer Residual network (ResNet) is utilized to classify multiple classes in eight SDD tasks, which include identification of scene levels, damage levels, and material types. The proposed ResNet achieved high accuracy for each task while the positions of the damage are not identifiable. In the second study, the existing ResNet and a segmentation network (U-Net) are combined into a new pipeline, cascaded networks, for categorizing and locating structural damage. The results show that the accuracy of damage detection is significantly improved compared to only using a segmentation network. In the third and fourth studies, end-to-end networks are developed and tested as a new solution to directly detect cracks and spalling in the image collections of recent large earthquakes. One of the proposed networks can achieve an accuracy above 67 .6% for all tested images at various scales and resolutions, and shows its robustness for these human-free detection tasks. As a preliminary field study, we applied the proposed method to detect damage in a concrete structure that was tested to study its progressive collapse performance. The experiments indicate that these solutions for automatic detection of structural damage using deep learning methods are feasible and promising. The training datasets and codes will be made available for the public upon the publication of this paper.

show abstract

A volumetric change detection framework using UAV oblique photogrammetry – a case study of ultra-high-resolution monitoring of progressive building collapse

Cited by 12 publications

References 28 publications

3D Change Detection Method for Exterior Wall of LNG Storage Tank Supported by Multi‐Source Spatial Data

3D Change Detection Method for Exterior Wall of LNG Storage Tank Supported by Multi‐Source Spatial Data

Large-scale DSM registration via motion averaging

Engineering deep learning methods on automatic detection of damage in infrastructure due to extreme events

Contact Info

Product

Resources

About