An Improved Automatic Pointwise Semantic Segmentation of a 3d Urban Scene From Mobile Terrestrial and Airborne Lidar Point Clouds: A Machine Learning Approach

Xing, Xufeng; Mostafavi, Mir Abolfazl; Edwards, Geoffrey; Sabo, N

doi:10.5194/isprs-annals-iv-4-w8-139-2019

Cited by 8 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The feature vector generally consists of quantitative values for each feature. For the task of extracting points belonging to our historical building from point clouds, semantic segmentation based on the feature vectors proposed in [38] is chosen to process the input point clouds presented in the previous section. In this work, among others, features such as (i) "directional height above" compares the height difference between a point and its neighbours in eight directions, (ii) "difference of normal" is based on a normal estimation which can better distinguish when vegetation and man-made objects are used.…”

Section: Semantic Segmentation Of Lidar Point Cloud For Object Extractionmentioning

confidence: 99%

Ontology-Based Semantic Conceptualisation of Historical Built Heritage to Generate Parametric Structured Models from Point Clouds

et al. 2021

View full text Add to dashboard Cite

Nowadays, cultural and historical built heritage can be more effectively preserved, valorised and documented using advanced geospatial technologies. In such a context, there is a major issue concerning the automation of the process and the extraction of useful information from a huge amount of spatial information acquired by means of advanced survey techniques (i.e., highly detailed LiDAR point clouds). In particular, in the case of historical built heritage (HBH) there are very few effective efforts. Therefore, in this paper, the focus is on establishing the connections between semantic and geometrical information in order to generate a parametric, structured model from point clouds using ontology as an effective approach for the formal conceptualisation of application domains. Hence, in this paper, an ontological schema is proposed to structure HBH representations, starting with international standards, vocabularies, and ontologies (CityGML-Geography Markup Language, International Committee for Documentation conceptual reference model (CIDOC-CRM), Industry Foundation Classes (IFC), Getty Art and Architecture Thesaurus (AAT), as well as reasoning about morphology of historical centres by analysis of real case studies) to represent the built and architecture domain. The validation of such schema is carried out by means of its use to guide the segmentation of a LiDAR point cloud from a castle, which is later used to generate parametric geometries to be used in a historical building information model (HBIM).

show abstract

Section: Semantic Segmentation Of Lidar Point Cloud For Object Extractionmentioning

confidence: 99%

Ontology-Based Semantic Conceptualisation of Historical Built Heritage to Generate Parametric Structured Models from Point Clouds

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Most of classification algorithms, including the Random Forest, rely on local and global descriptors, computed at a point scale, or at a global scale (Hackel et al, 2016;Xing et al, 2019). Such descriptors are usually easy and fast to compute, while providing meaningfull geometrical information.…”

Section: Descriptors Computationmentioning

confidence: 99%

Fast Weakly Supervised Detection of Railway-Related Infrastructures in Lidar Acquisitions

Guinard

Riant

Michelin

et al. 2021

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

View full text Add to dashboard Cite

Abstract. Railroad environments are peculiar, as they combine dense urban areas, along with rural parts. They also display a very specific spatial organization. In order to monitor a railway network a at country scale, LiDAR sensors can be equipped on a running train, performing a full acquisition of the network. Then most processing steps are manually done. In this paper, we propose to improve performances and production flow by creating a classification of the acquired data. However, there exists no public benchmark, and little work on LiDAR data classification in railroad environments. Thus, we propose a weakly supervised method for the pointwise classification of such data. We show that our method can be improved by using the l0-cut pursuit algorithm and regularize the noisy pointwise classification on the produced segmentation. As production is envisaged in our context, we designed our implementation such that it is computationally efficient. We evaluate our results against a manual classification, and show that our method can reach a FScore of 0.96 with just a few samples of each class.

show abstract

“…The proposed method results in the main overall accuracy of 95%. More recently, Xing et al used the Hackel et al workflow as a basis and developed a more robust workflow by adding a series of features computed based on the difference of normal vectors for better identification [19,20]. Their study demonstrated a 2% improvement on average.…”

Section: Studies Used 3d Point Clouds For Detection and Classificationmentioning

confidence: 99%

Deep Learning Classification of 2D Orthomosaic Images and 3D Point Clouds for Post-Event Structural Damage Assessment

Yi-jun

Mohammadi

Wood

2020

Drones

View full text Add to dashboard Cite

Efficient and rapid data collection techniques are necessary to obtain transitory information in the aftermath of natural hazards, which is not only useful for post-event management and planning, but also for post-event structural damage assessment. Aerial imaging from unpiloted (gender-neutral, but also known as unmanned) aerial systems (UASs) or drones permits highly detailed site characterization, in particular in the aftermath of extreme events with minimal ground support, to document current conditions of the region of interest. However, aerial imaging results in a massive amount of data in the form of two-dimensional (2D) orthomosaic images and three-dimensional (3D) point clouds. Both types of datasets require effective and efficient data processing workflows to identify various damage states of structures. This manuscript aims to introduce two deep learning models based on both 2D and 3D convolutional neural networks to process the orthomosaic images and point clouds, for post windstorm classification. In detail, 2D convolutional neural networks (2D CNN) are developed based on transfer learning from two well-known networks AlexNet and VGGNet. In contrast, a 3D fully convolutional network (3DFCN) with skip connections was developed and trained based on the available point cloud data. Within this study, the datasets were created based on data from the aftermath of Hurricanes Harvey (Texas) and Maria (Puerto Rico). The developed 2DCNN and 3DFCN models were compared quantitatively based on the performance measures, and it was observed that the 3DFCN was more robust in detecting the various classes. This demonstrates the value and importance of 3D datasets, particularly the depth information, to distinguish between instances that represent different damage states in structures.

show abstract

An Improved Automatic Pointwise Semantic Segmentation of a 3d Urban Scene From Mobile Terrestrial and Airborne Lidar Point Clouds: A Machine Learning Approach

Cited by 8 publications

References 25 publications

Ontology-Based Semantic Conceptualisation of Historical Built Heritage to Generate Parametric Structured Models from Point Clouds

Ontology-Based Semantic Conceptualisation of Historical Built Heritage to Generate Parametric Structured Models from Point Clouds

Fast Weakly Supervised Detection of Railway-Related Infrastructures in Lidar Acquisitions

Deep Learning Classification of 2D Orthomosaic Images and 3D Point Clouds for Post-Event Structural Damage Assessment

Contact Info

Product

Resources

About