Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis

Yang Xiao; Liam Daniel; Marina Gashinova

doi:10.1109/RadarConf2043947.2020.9266596

Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis

Yang Xiao, Liam Daniel, Marina Gashinova

Electronic, Electrical and Systems Engineering

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

Abstract

Segmentation and potential classification of surface and obstacle regions in automotive radar imagery is the key enabler of effective path planning in autonomous driving. As opposed to traditional radar processing where clutter is considered as an unwanted return and should be effectively removed, autonomous driving requires full scene assessment, where clutter carries necessary information for situational awareness of the autonomous platform and needs to be fully assessed to find the passable areas. In this paper, the statistical distribution features of the radar intensity data of several road-related scenes including asphalt, grass, shadow and target object areas are investigated. The algorithm of classification is developed based on distribution feature extraction and a multivariate Gaussian distribution (MGD) model. Under test dataset recorded by multi-sensor suit was used to evaluate the confusion matrix and F1 score of this classification algorithm.

Original language	English
Title of host publication	2020 IEEE Radar Conference, RadarConf 2020
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
ISBN (Electronic)	9781728189420
DOIs	https://doi.org/10.1109/RadarConf2043947.2020.9266596
Publication status	Published - 21 Sept 2020
Event	2020 IEEE Radar Conference, RadarConf 2020 - Florence, Italy Duration: 21 Sept 2020 → 25 Sept 2020

Publication series

Name	IEEE National Radar Conference - Proceedings
Volume	2020-September
ISSN (Print)	1097-5659

Conference

Conference	2020 IEEE Radar Conference, RadarConf 2020
Country/Territory	Italy
City	Florence
Period	21/09/20 → 25/09/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Keywords

Automotive sensor
image segmentation
multivariate Gaussian distribution
radar imaging
statistical distribution feature extraction
surface classification

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1109/RadarConf2043947.2020.9266596

Cite this

Xiao, Y., Daniel, L., & Gashinova, M. (2020). Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis. In 2020 IEEE Radar Conference, RadarConf 2020 Article 9266596 (IEEE National Radar Conference - Proceedings; Vol. 2020-September). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/RadarConf2043947.2020.9266596

@inproceedings{b4b38160b0c5437394de807e7c45b349,

title = "Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis",

abstract = "Segmentation and potential classification of surface and obstacle regions in automotive radar imagery is the key enabler of effective path planning in autonomous driving. As opposed to traditional radar processing where clutter is considered as an unwanted return and should be effectively removed, autonomous driving requires full scene assessment, where clutter carries necessary information for situational awareness of the autonomous platform and needs to be fully assessed to find the passable areas. In this paper, the statistical distribution features of the radar intensity data of several road-related scenes including asphalt, grass, shadow and target object areas are investigated. The algorithm of classification is developed based on distribution feature extraction and a multivariate Gaussian distribution (MGD) model. Under test dataset recorded by multi-sensor suit was used to evaluate the confusion matrix and F1 score of this classification algorithm.",

keywords = "Automotive sensor, image segmentation, multivariate Gaussian distribution, radar imaging, statistical distribution feature extraction, surface classification",

author = "Yang Xiao and Liam Daniel and Marina Gashinova",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE Radar Conference, RadarConf 2020 ; Conference date: 21-09-2020 Through 25-09-2020",

year = "2020",

month = sep,

day = "21",

doi = "10.1109/RadarConf2043947.2020.9266596",

language = "English",

series = "IEEE National Radar Conference - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

booktitle = "2020 IEEE Radar Conference, RadarConf 2020",

}

Xiao, Y, Daniel, L & Gashinova, M 2020, Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis. in 2020 IEEE Radar Conference, RadarConf 2020., 9266596, IEEE National Radar Conference - Proceedings, vol. 2020-September, Institute of Electrical and Electronics Engineers (IEEE), 2020 IEEE Radar Conference, RadarConf 2020, Florence, Italy, 21/09/20. https://doi.org/10.1109/RadarConf2043947.2020.9266596

Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis. / Xiao, Yang; Daniel, Liam ; Gashinova, Marina.
2020 IEEE Radar Conference, RadarConf 2020. Institute of Electrical and Electronics Engineers (IEEE), 2020. 9266596 (IEEE National Radar Conference - Proceedings; Vol. 2020-September).

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

TY - GEN

T1 - Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis

AU - Xiao, Yang

AU - Daniel, Liam

AU - Gashinova, Marina

PY - 2020/9/21

Y1 - 2020/9/21

N2 - Segmentation and potential classification of surface and obstacle regions in automotive radar imagery is the key enabler of effective path planning in autonomous driving. As opposed to traditional radar processing where clutter is considered as an unwanted return and should be effectively removed, autonomous driving requires full scene assessment, where clutter carries necessary information for situational awareness of the autonomous platform and needs to be fully assessed to find the passable areas. In this paper, the statistical distribution features of the radar intensity data of several road-related scenes including asphalt, grass, shadow and target object areas are investigated. The algorithm of classification is developed based on distribution feature extraction and a multivariate Gaussian distribution (MGD) model. Under test dataset recorded by multi-sensor suit was used to evaluate the confusion matrix and F1 score of this classification algorithm.

AB - Segmentation and potential classification of surface and obstacle regions in automotive radar imagery is the key enabler of effective path planning in autonomous driving. As opposed to traditional radar processing where clutter is considered as an unwanted return and should be effectively removed, autonomous driving requires full scene assessment, where clutter carries necessary information for situational awareness of the autonomous platform and needs to be fully assessed to find the passable areas. In this paper, the statistical distribution features of the radar intensity data of several road-related scenes including asphalt, grass, shadow and target object areas are investigated. The algorithm of classification is developed based on distribution feature extraction and a multivariate Gaussian distribution (MGD) model. Under test dataset recorded by multi-sensor suit was used to evaluate the confusion matrix and F1 score of this classification algorithm.

KW - Automotive sensor

KW - image segmentation

KW - multivariate Gaussian distribution

KW - radar imaging

KW - statistical distribution feature extraction

KW - surface classification

UR - http://www.scopus.com/inward/record.url?scp=85098566009&partnerID=8YFLogxK

U2 - 10.1109/RadarConf2043947.2020.9266596

DO - 10.1109/RadarConf2043947.2020.9266596

M3 - Conference contribution

AN - SCOPUS:85098566009

T3 - IEEE National Radar Conference - Proceedings

BT - 2020 IEEE Radar Conference, RadarConf 2020

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 2020 IEEE Radar Conference, RadarConf 2020

Y2 - 21 September 2020 through 25 September 2020

ER -

Feature-based Classification for Image Segmentation in Automotive Radar Based on Statistical Distribution Analysis

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this