Damaged lane markings detection method with label propagation

Tetsuo Nukita, Yasunari Kishimoto, Yasuhiro Iida, Makoto Kawano, Takuro Yonezawa, Jin Nakazawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose a damaged traffic lane detection method ensuring high accuracy in spite of only a few number of supervised data which are labeled traffic lane images. In general, supervised machine learning approach is very powerful for image classification. However, preparing a large amount of supervised data is time-consuming task because labeling damaged or not damaged is usually done manually through visual inspection of images. Thus, lowering the cost of labeling data is a great concern. To this end, we adopt a semi-supervised machine learning approach which learns from both labeled and unlabeled data by constructing graph based on the image similarity. We captured a large amount of the road lane images. Then, we constructed graph structure whose nodes are the road lane images and whose edges are the similarity between the images. In several nodes, we assigned labels which denote "damaged" or "not damaged." Finally we utilized the label propagation, which made it possible to infer the labels of the unlabeled data from the labeled data. These estimation resulted in the accuracy rate over 85% from the supervised data, which accounted for only 1.8% of the total data.

Original languageEnglish
Title of host publicationProceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages203-208
Number of pages6
ISBN (Electronic)9781538677599
DOIs
Publication statusPublished - 2019 Jan 9
Event24th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018 - Hakodate, Japan
Duration: 2018 Aug 292018 Aug 31

Publication series

NameProceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018

Conference

Conference24th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018
CountryJapan
CityHakodate
Period18/8/2918/8/31

Fingerprint

Labels
Labeling
Learning systems
Image classification
Inspection
Costs

Keywords

  • Image detection
  • Label propagation
  • Supervised machine learning

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications

Cite this

Nukita, T., Kishimoto, Y., Iida, Y., Kawano, M., Yonezawa, T., & Nakazawa, J. (2019). Damaged lane markings detection method with label propagation. In Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018 (pp. 203-208). [8607250] (Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RTCSA.2018.00032

Damaged lane markings detection method with label propagation. / Nukita, Tetsuo; Kishimoto, Yasunari; Iida, Yasuhiro; Kawano, Makoto; Yonezawa, Takuro; Nakazawa, Jin.

Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018. Institute of Electrical and Electronics Engineers Inc., 2019. p. 203-208 8607250 (Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nukita, T, Kishimoto, Y, Iida, Y, Kawano, M, Yonezawa, T & Nakazawa, J 2019, Damaged lane markings detection method with label propagation. in Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018., 8607250, Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018, Institute of Electrical and Electronics Engineers Inc., pp. 203-208, 24th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018, Hakodate, Japan, 18/8/29. https://doi.org/10.1109/RTCSA.2018.00032
Nukita T, Kishimoto Y, Iida Y, Kawano M, Yonezawa T, Nakazawa J. Damaged lane markings detection method with label propagation. In Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018. Institute of Electrical and Electronics Engineers Inc. 2019. p. 203-208. 8607250. (Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018). https://doi.org/10.1109/RTCSA.2018.00032
Nukita, Tetsuo ; Kishimoto, Yasunari ; Iida, Yasuhiro ; Kawano, Makoto ; Yonezawa, Takuro ; Nakazawa, Jin. / Damaged lane markings detection method with label propagation. Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 203-208 (Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018).
@inproceedings{6322f08d5b6f42cd829926b7b015b337,
title = "Damaged lane markings detection method with label propagation",
abstract = "We propose a damaged traffic lane detection method ensuring high accuracy in spite of only a few number of supervised data which are labeled traffic lane images. In general, supervised machine learning approach is very powerful for image classification. However, preparing a large amount of supervised data is time-consuming task because labeling damaged or not damaged is usually done manually through visual inspection of images. Thus, lowering the cost of labeling data is a great concern. To this end, we adopt a semi-supervised machine learning approach which learns from both labeled and unlabeled data by constructing graph based on the image similarity. We captured a large amount of the road lane images. Then, we constructed graph structure whose nodes are the road lane images and whose edges are the similarity between the images. In several nodes, we assigned labels which denote {"}damaged{"} or {"}not damaged.{"} Finally we utilized the label propagation, which made it possible to infer the labels of the unlabeled data from the labeled data. These estimation resulted in the accuracy rate over 85{\%} from the supervised data, which accounted for only 1.8{\%} of the total data.",
keywords = "Image detection, Label propagation, Supervised machine learning",
author = "Tetsuo Nukita and Yasunari Kishimoto and Yasuhiro Iida and Makoto Kawano and Takuro Yonezawa and Jin Nakazawa",
year = "2019",
month = "1",
day = "9",
doi = "10.1109/RTCSA.2018.00032",
language = "English",
series = "Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "203--208",
booktitle = "Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018",

}

TY - GEN

T1 - Damaged lane markings detection method with label propagation

AU - Nukita, Tetsuo

AU - Kishimoto, Yasunari

AU - Iida, Yasuhiro

AU - Kawano, Makoto

AU - Yonezawa, Takuro

AU - Nakazawa, Jin

PY - 2019/1/9

Y1 - 2019/1/9

N2 - We propose a damaged traffic lane detection method ensuring high accuracy in spite of only a few number of supervised data which are labeled traffic lane images. In general, supervised machine learning approach is very powerful for image classification. However, preparing a large amount of supervised data is time-consuming task because labeling damaged or not damaged is usually done manually through visual inspection of images. Thus, lowering the cost of labeling data is a great concern. To this end, we adopt a semi-supervised machine learning approach which learns from both labeled and unlabeled data by constructing graph based on the image similarity. We captured a large amount of the road lane images. Then, we constructed graph structure whose nodes are the road lane images and whose edges are the similarity between the images. In several nodes, we assigned labels which denote "damaged" or "not damaged." Finally we utilized the label propagation, which made it possible to infer the labels of the unlabeled data from the labeled data. These estimation resulted in the accuracy rate over 85% from the supervised data, which accounted for only 1.8% of the total data.

AB - We propose a damaged traffic lane detection method ensuring high accuracy in spite of only a few number of supervised data which are labeled traffic lane images. In general, supervised machine learning approach is very powerful for image classification. However, preparing a large amount of supervised data is time-consuming task because labeling damaged or not damaged is usually done manually through visual inspection of images. Thus, lowering the cost of labeling data is a great concern. To this end, we adopt a semi-supervised machine learning approach which learns from both labeled and unlabeled data by constructing graph based on the image similarity. We captured a large amount of the road lane images. Then, we constructed graph structure whose nodes are the road lane images and whose edges are the similarity between the images. In several nodes, we assigned labels which denote "damaged" or "not damaged." Finally we utilized the label propagation, which made it possible to infer the labels of the unlabeled data from the labeled data. These estimation resulted in the accuracy rate over 85% from the supervised data, which accounted for only 1.8% of the total data.

KW - Image detection

KW - Label propagation

KW - Supervised machine learning

UR - http://www.scopus.com/inward/record.url?scp=85061769085&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061769085&partnerID=8YFLogxK

U2 - 10.1109/RTCSA.2018.00032

DO - 10.1109/RTCSA.2018.00032

M3 - Conference contribution

T3 - Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018

SP - 203

EP - 208

BT - Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -