A Review Of
Point Clouds Segmentation and Classification Algorithms

Grilli, Eleonora; Menna, Fabio; Remondino, Fabio

doi:10.5194/isprs-archives-xlii-2-w3-339-2017

Cited by 314 publications

(190 citation statements)

References 35 publications

Supporting

Mentioning

183

Contrasting

Unclassified

Order By: Relevance

“…It is also possible to generate simple handcrafted features such as sparsity, planarity, linearity, and many more [13]. Incorporating those features with our fusion approach could potentially improve the performance of a resulting classifier.…”

Section: Interesting Insightsmentioning

confidence: 99%

See 1 more Smart Citation

Land Cover Segmentation of Airborne LiDAR Data Using Stochastic Atrous Network

et al. 2018

View full text Add to dashboard Cite

Inspired by the success of deep learning techniques in dense-label prediction and the increasing availability of high precision airborne light detection and ranging (LiDAR) data, we present a research process that compares a collection of well-proven semantic segmentation architectures based on the deep learning approach. Our investigation concludes with the proposition of some novel deep learning architectures for generating detailed land resource maps by employing a semantic segmentation approach. The contribution of our work is threefold.(1) First, we implement the multiclass version of the intersection-over-union (IoU) loss function that contributes to handling highly imbalanced datasets and preventing overfitting. (2) Thereafter, we propose a novel deep learning architecture integrating the deep atrous network architecture with the stochastic depth approach for speeding up the learning process, and impose a regularization effect. (3) Finally, we introduce an early fusion deep layer that combines image-based and LiDAR-derived features. In a benchmark study carried out using the Follo 2014 LiDAR data and the NIBIO AR5 land resources dataset, we compare our proposals to other deep learning architectures. A quantitative comparison shows that our best proposal provides more than 5% relative improvement in terms of mean intersection-over-union over the atrous network, providing a basis for a more frequent and improved use of LiDAR data for automatic land cover segmentation.

show abstract

Section: Interesting Insightsmentioning

confidence: 99%

“…However, such approaches limit the ability to provide a richer representation of the data [13]. The chosen features may not be sufficient to characterize the uniqueness of a certain class or object [14], and the quality of the resulting classifier depends heavily on the feature engineering output.…”

Section: Introductionmentioning

confidence: 99%

Land Cover Segmentation of Airborne LiDAR Data Using Stochastic Atrous Network

et al. 2018

View full text Add to dashboard Cite

show abstract

“…It is based on assigning to each point of the cloud the proper attribute associated with the object on which the laser beam reflected. This way points lying on the ground, representing low vegetation, medium and high buildings and other classes from the whole cloud can be determined (Grilli et al, 2017). In the case of LIDAR data, it was necessary to establish the characteristics of each of the classes to which the individual elements of the terrain belong (Table 2).…”

Section: Classification and Filtrationmentioning

confidence: 99%

Effective Detection of Sub-Surface Archeological Features From Laser Scanning Point Clouds and Imagery Data

Fryśkowska¹,

Kędzierski²,

Walczykowski³

et al. 2017

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:The archaeological heritage is non-renewable, and any invasive research or other actions leading to the intervention of mechanical or chemical into the ground lead to the destruction of the archaeological site in whole or in part. For this reason, modern archeology is looking for alternative methods of non-destructive and non-invasive methods of new objects identification. The concept of aerial archeology is relation between the presence of the archaeological site in the particular localization, and the phenomena that in the same place can be observed on the terrain surface form airborne platform. One of the most appreciated, moreover, extremely precise, methods of such measurements is airborne laser scanning. In research airborne laser scanning point cloud with a density of 5 points/sq. m was used. Additionally unmanned aerial vehicle imagery data was acquired. Test area is located in central Europe. The preliminary verification of potentially microstructures localization was the creation of digital terrain and surface models. These models gave an information about the differences in elevation, as well as regular shapes and sizes that can be related to the former settlement/sub-surface feature. The paper presents the results of the detection of potentially sub-surface microstructure fields in the forestry area.

show abstract

“…Methods based on the exploitation of the local geometric structure have been studied by multiple researchers. Recent overviews are provided by Guan et al (2016), Grilli et al (2017) and Lemmens (2017).…”

Section: Introductionmentioning

confidence: 99%

Classification of Mobile Laser Scanning Point Clouds From Height Features

Zheng

Lemmens

Oosterom

2017

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:The demand for 3D maps of cities and road networks is steadily growing and mobile laser scanning (MLS) systems are often the preferred geo-data acquisition method for capturing such scenes. Because MLS systems are mounted on cars or vans they can acquire billions of points of road scenes within a few hours of survey. Manual processing of point clouds is labour intensive and thus time consuming and expensive. Hence, the need for rapid and automated methods for 3D mapping of dense point clouds is growing exponentially. The last five years the research on automated 3D mapping of MLS data has tremendously intensified. In this paper, we present our work on automated classification of MLS point clouds. In the present stage of the research we exploited three featurestwo height components and one reflectance value, and achieved an overall accuracy of 73%, which is really encouraging for further refining our approach.

show abstract

A Review Of Point Clouds Segmentation and Classification Algorithms

Cited by 314 publications

References 35 publications

Land Cover Segmentation of Airborne LiDAR Data Using Stochastic Atrous Network

Land Cover Segmentation of Airborne LiDAR Data Using Stochastic Atrous Network

Effective Detection of Sub-Surface Archeological Features From Laser Scanning Point Clouds and Imagery Data

Classification of Mobile Laser Scanning Point Clouds From Height Features

Contact Info

Product

Resources

About