A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV

Jinbo Zhang; Yutong Zhang; Ding Yan; Jiayuan Shan; Zhaodi Wang; Pingping Wu

doi:10.1145/3448734.3450817

A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV

Jinbo Zhang, Yutong Zhang, Ding Yan^*, Jiayuan Shan, Zhaodi Wang, Pingping Wu

^*Corresponding author for this work

School of Aerospace Engineering

Beijing Institute of Technology

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

1 Citation (Scopus)

Abstract

Fast size measurement is widely used in many fields, such as airport security inspection, logistics classification and so on. In order to reduce the measurement error and enhance the measurement speed, we design a fast size measurement system based on multiple RGB-D cameras and a main axis calculation mechanism based on the distribution of pointcloud normal vectors (MA-PNV) is proposed. In our algorithm, the pointcloud of the object is acquired through coordinate transformation and background subtraction. After statistical outlier removal, the normal vectors of the pointcloud is obtained by the local surface fitting. Then the main axis is calculated based on MA-PNV and size measurement of the object is accomplished. Experimental evaluation is carried out on objects with different shapes, colors and under different illumination conditions. The results illustrate that our algorithm is effective for fast size measurement.

Original language	English
Title of host publication	Proceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021
Publisher	Association for Computing Machinery
ISBN (Electronic)	9781450389570
DOIs	https://doi.org/10.1145/3448734.3450817
Publication status	Published - 28 Jan 2021
Event	2nd International Conference on Computing and Data Science, CONF-CDS 2021 - Stanford, United States Duration: 28 Jan 2021 → 30 Jan 2021

Publication series

Name	ACM International Conference Proceeding Series
Volume	PartF168982

Conference

Conference	2nd International Conference on Computing and Data Science, CONF-CDS 2021
Country/Territory	United States
City	Stanford
Period	28/01/21 → 30/01/21

Keywords

MA-PNV
RGB-D cameras
Size measurement
pointcloud stitching

Access to Document

10.1145/3448734.3450817

Cite this

Zhang, J., Zhang, Y., Yan, D., Shan, J., Wang, Z., & Wu, P. (2021). A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV. In Proceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021 Article 3450817 (ACM International Conference Proceeding Series; Vol. PartF168982). Association for Computing Machinery. https://doi.org/10.1145/3448734.3450817

@inproceedings{cefe3d5479c54972a70a9f80148a59f2,

title = "A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV",

abstract = "Fast size measurement is widely used in many fields, such as airport security inspection, logistics classification and so on. In order to reduce the measurement error and enhance the measurement speed, we design a fast size measurement system based on multiple RGB-D cameras and a main axis calculation mechanism based on the distribution of pointcloud normal vectors (MA-PNV) is proposed. In our algorithm, the pointcloud of the object is acquired through coordinate transformation and background subtraction. After statistical outlier removal, the normal vectors of the pointcloud is obtained by the local surface fitting. Then the main axis is calculated based on MA-PNV and size measurement of the object is accomplished. Experimental evaluation is carried out on objects with different shapes, colors and under different illumination conditions. The results illustrate that our algorithm is effective for fast size measurement.",

keywords = "MA-PNV, RGB-D cameras, Size measurement, pointcloud stitching",

author = "Jinbo Zhang and Yutong Zhang and Ding Yan and Jiayuan Shan and Zhaodi Wang and Pingping Wu",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 2nd International Conference on Computing and Data Science, CONF-CDS 2021 ; Conference date: 28-01-2021 Through 30-01-2021",

year = "2021",

month = jan,

day = "28",

doi = "10.1145/3448734.3450817",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

booktitle = "Proceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021",

}

Zhang, J, Zhang, Y, Yan, D , Shan, J, Wang, Z & Wu, P 2021, A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV. in Proceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021., 3450817, ACM International Conference Proceeding Series, vol. PartF168982, Association for Computing Machinery, 2nd International Conference on Computing and Data Science, CONF-CDS 2021, Stanford, United States, 28/01/21. https://doi.org/10.1145/3448734.3450817

A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV. / Zhang, Jinbo; Zhang, Yutong; Yan, Ding et al.
Proceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021. Association for Computing Machinery, 2021. 3450817 (ACM International Conference Proceeding Series; Vol. PartF168982).

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

TY - GEN

T1 - A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV

AU - Zhang, Jinbo

AU - Zhang, Yutong

AU - Yan, Ding

AU - Shan, Jiayuan

AU - Wang, Zhaodi

AU - Wu, Pingping

PY - 2021/1/28

Y1 - 2021/1/28

N2 - Fast size measurement is widely used in many fields, such as airport security inspection, logistics classification and so on. In order to reduce the measurement error and enhance the measurement speed, we design a fast size measurement system based on multiple RGB-D cameras and a main axis calculation mechanism based on the distribution of pointcloud normal vectors (MA-PNV) is proposed. In our algorithm, the pointcloud of the object is acquired through coordinate transformation and background subtraction. After statistical outlier removal, the normal vectors of the pointcloud is obtained by the local surface fitting. Then the main axis is calculated based on MA-PNV and size measurement of the object is accomplished. Experimental evaluation is carried out on objects with different shapes, colors and under different illumination conditions. The results illustrate that our algorithm is effective for fast size measurement.

AB - Fast size measurement is widely used in many fields, such as airport security inspection, logistics classification and so on. In order to reduce the measurement error and enhance the measurement speed, we design a fast size measurement system based on multiple RGB-D cameras and a main axis calculation mechanism based on the distribution of pointcloud normal vectors (MA-PNV) is proposed. In our algorithm, the pointcloud of the object is acquired through coordinate transformation and background subtraction. After statistical outlier removal, the normal vectors of the pointcloud is obtained by the local surface fitting. Then the main axis is calculated based on MA-PNV and size measurement of the object is accomplished. Experimental evaluation is carried out on objects with different shapes, colors and under different illumination conditions. The results illustrate that our algorithm is effective for fast size measurement.

KW - MA-PNV

KW - RGB-D cameras

KW - Size measurement

KW - pointcloud stitching

UR - http://www.scopus.com/inward/record.url?scp=85105996983&partnerID=8YFLogxK

U2 - 10.1145/3448734.3450817

DO - 10.1145/3448734.3450817

M3 - Conference contribution

AN - SCOPUS:85105996983

T3 - ACM International Conference Proceeding Series

BT - Proceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021

PB - Association for Computing Machinery

T2 - 2nd International Conference on Computing and Data Science, CONF-CDS 2021

Y2 - 28 January 2021 through 30 January 2021

ER -

A Size Measurement Algorithm Based on Multiple RGB-D Cameras and MA-PNV

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this