3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data

Nur Hamid; Ari Wibisono; M. Anwar Ma'Sum; Ahmad Gamal; Roni Ardhianto; Aniati Murni Arymurthy; Wisnu Jatmiko

doi:10.1109/ACIRS.2019.8935980

3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data

Nur Hamid
, Ari Wibisono
, M. Anwar Ma'Sum
, Ahmad Gamal
, Roni Ardhianto
, Aniati Murni Arymurthy
, Wisnu Jatmiko

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

3-dimensional data contains more informative visualization than a 2-dimensional one. LiDAR sensor produces 3D data or point cloud data. There have been many implementations of LiDAR data such as for building detection, urban area modeling, and land cover analysis. This study will analyze land cover because of its substantial benefits. The purpose of this study is to produce semantic segmentation of land cover from LiDAR data by implementing EdgeConv Algorithm from Dynamic Graph Convolutional Neural Network (DGCNN). The dataset in this study is LiDAR data of Kupang, one of the areas in Indonesia. This work achieves the average accuracy of 67.76% for DGCNN better than the state-of-the-art method PointNet (previous method) with 64.97% by implementing the point cloud dataset from LiDAR data of Kupang. This is because the edge convolution could recognize the global shape structure and local neighborhood information so that it could improve the segmentation performance result.

Original language	English
Title of host publication	2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	216-220
Number of pages	5
ISBN (Electronic)	9781728122298
DOIs	https://doi.org/10.1109/ACIRS.2019.8935980
State	Published - Jul 2019
Externally published	Yes
Event	4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019 - Nagoya, Japan Duration: 13 Jul 2019 → 15 Jul 2019

Publication series

Name	2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019

Conference

Conference	4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019
Country/Territory	Japan
City	Nagoya
Period	13/07/19 → 15/07/19

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Keywords

3-dimensional edge detection
3D CNN
edge convolution
land cover
lidar
semantic segmentation

ASJC Scopus subject areas

Artificial Intelligence
Information Systems
Control and Optimization

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ACIRS.2019.8935980

Cite this

Hamid, N., Wibisono, A., Anwar Ma'Sum, M., Gamal, A., Ardhianto, R., Arymurthy, A. M., & Jatmiko, W. (2019). 3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data. In 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019 (pp. 216-220). Article 8935980 (2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACIRS.2019.8935980

Hamid, Nur ; Wibisono, Ari ; Anwar Ma'Sum, M. et al. / 3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data. 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 216-220 (2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019).

@inproceedings{41632e1c7ddd45bf96f54ca5d34927ac,

title = "3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data",

abstract = "3-dimensional data contains more informative visualization than a 2-dimensional one. LiDAR sensor produces 3D data or point cloud data. There have been many implementations of LiDAR data such as for building detection, urban area modeling, and land cover analysis. This study will analyze land cover because of its substantial benefits. The purpose of this study is to produce semantic segmentation of land cover from LiDAR data by implementing EdgeConv Algorithm from Dynamic Graph Convolutional Neural Network (DGCNN). The dataset in this study is LiDAR data of Kupang, one of the areas in Indonesia. This work achieves the average accuracy of 67.76\% for DGCNN better than the state-of-the-art method PointNet (previous method) with 64.97\% by implementing the point cloud dataset from LiDAR data of Kupang. This is because the edge convolution could recognize the global shape structure and local neighborhood information so that it could improve the segmentation performance result.",

keywords = "3-dimensional edge detection, 3D CNN, edge convolution, land cover, lidar, semantic segmentation",

author = "Nur Hamid and Ari Wibisono and \{Anwar Ma'Sum\}, M. and Ahmad Gamal and Roni Ardhianto and Arymurthy, \{Aniati Murni\} and Wisnu Jatmiko",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019 ; Conference date: 13-07-2019 Through 15-07-2019",

year = "2019",

month = jul,

doi = "10.1109/ACIRS.2019.8935980",

language = "English",

series = "2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "216--220",

booktitle = "2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019",

address = "United States",

}

Hamid, N, Wibisono, A, Anwar Ma'Sum, M, Gamal, A, Ardhianto, R, Arymurthy, AM & Jatmiko, W 2019, 3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data. in 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019., 8935980, 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019, Institute of Electrical and Electronics Engineers Inc., pp. 216-220, 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019, Nagoya, Japan, 13/07/19. https://doi.org/10.1109/ACIRS.2019.8935980

3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data. / Hamid, Nur; Wibisono, Ari; Anwar Ma'Sum, M. et al.
2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 216-220 8935980 (2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - 3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data

AU - Hamid, Nur

AU - Wibisono, Ari

AU - Anwar Ma'Sum, M.

AU - Gamal, Ahmad

AU - Ardhianto, Roni

AU - Arymurthy, Aniati Murni

AU - Jatmiko, Wisnu

PY - 2019/7

Y1 - 2019/7

N2 - 3-dimensional data contains more informative visualization than a 2-dimensional one. LiDAR sensor produces 3D data or point cloud data. There have been many implementations of LiDAR data such as for building detection, urban area modeling, and land cover analysis. This study will analyze land cover because of its substantial benefits. The purpose of this study is to produce semantic segmentation of land cover from LiDAR data by implementing EdgeConv Algorithm from Dynamic Graph Convolutional Neural Network (DGCNN). The dataset in this study is LiDAR data of Kupang, one of the areas in Indonesia. This work achieves the average accuracy of 67.76% for DGCNN better than the state-of-the-art method PointNet (previous method) with 64.97% by implementing the point cloud dataset from LiDAR data of Kupang. This is because the edge convolution could recognize the global shape structure and local neighborhood information so that it could improve the segmentation performance result.

AB - 3-dimensional data contains more informative visualization than a 2-dimensional one. LiDAR sensor produces 3D data or point cloud data. There have been many implementations of LiDAR data such as for building detection, urban area modeling, and land cover analysis. This study will analyze land cover because of its substantial benefits. The purpose of this study is to produce semantic segmentation of land cover from LiDAR data by implementing EdgeConv Algorithm from Dynamic Graph Convolutional Neural Network (DGCNN). The dataset in this study is LiDAR data of Kupang, one of the areas in Indonesia. This work achieves the average accuracy of 67.76% for DGCNN better than the state-of-the-art method PointNet (previous method) with 64.97% by implementing the point cloud dataset from LiDAR data of Kupang. This is because the edge convolution could recognize the global shape structure and local neighborhood information so that it could improve the segmentation performance result.

KW - 3-dimensional edge detection

KW - 3D CNN

KW - edge convolution

KW - land cover

KW - lidar

KW - semantic segmentation

UR - https://www.scopus.com/pages/publications/85078064012

U2 - 10.1109/ACIRS.2019.8935980

DO - 10.1109/ACIRS.2019.8935980

M3 - Conference contribution

AN - SCOPUS:85078064012

T3 - 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019

SP - 216

EP - 220

BT - 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019

Y2 - 13 July 2019 through 15 July 2019

ER -

Hamid N, Wibisono A, Anwar Ma'Sum M, Gamal A, Ardhianto R, Arymurthy AM et al. 3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data. In 2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 216-220. 8935980. (2019 4th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2019). doi: 10.1109/ACIRS.2019.8935980

3D Edge Convolution in Deep Neural Network Implementation for Land Cover Semantic Segmentation of Airborne LiDAR Data

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Fingerprint

Cite this