Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network
In this paper, a new algorithm for extracting the laser fringe center is proposed. Based on a deep learning skeleton extraction network, the laser stripe center can be extracted quickly and accurately. Skeleton extraction is the process of reducing the shape image to its approximate central axis rep...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/3/859 |
| _version_ | 1797484824282267648 |
|---|---|
| author | Shuai Luo Kai Yang Lijuan Yang Yong Wang Xiaorong Gao Tianci Jiang Chunjiang Li |
| author_facet | Shuai Luo Kai Yang Lijuan Yang Yong Wang Xiaorong Gao Tianci Jiang Chunjiang Li |
| author_sort | Shuai Luo |
| collection | DOAJ |
| description | In this paper, a new algorithm for extracting the laser fringe center is proposed. Based on a deep learning skeleton extraction network, the laser stripe center can be extracted quickly and accurately. Skeleton extraction is the process of reducing the shape image to its approximate central axis representation while maintaining the image’s topological and geometric shape. Skeleton extraction is an important step in topological and geometric shape analysis. According to the characteristics of the wheelset laser curve dataset, a new skeleton extraction network, a hierarchical skeleton network (LuoNet), is proposed. The proposed architecture has three levels of the encoder–decoder network, and YE Module interconnection is designed between each level of the encoder and decoder network. In the wheelset laser curve dataset, the F1_score can reach 0.714. Compared with the traditional laser curve center extraction algorithm, the proposed LuoNet algorithm has the advantages of short running time, high accuracy, and stable extraction results. |
| first_indexed | 2024-03-09T23:10:03Z |
| format | Article |
| id | doaj.art-aed6b1e062c643aea322072ecefec58c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T23:10:03Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-aed6b1e062c643aea322072ecefec58c2023-11-23T17:46:34ZengMDPI AGSensors1424-82202022-01-0122385910.3390/s22030859Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction NetworkShuai Luo0Kai Yang1Lijuan Yang2Yong Wang3Xiaorong Gao4Tianci Jiang5Chunjiang Li6School of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Mathematics, Sichuan Normal University, Chengdu 610066, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Mechanical Engineering, Waseda University, Kitakyushu 8080135, JapanSchool of Mechanics, Zhejiang University, Hangzhou 310058, ChinaIn this paper, a new algorithm for extracting the laser fringe center is proposed. Based on a deep learning skeleton extraction network, the laser stripe center can be extracted quickly and accurately. Skeleton extraction is the process of reducing the shape image to its approximate central axis representation while maintaining the image’s topological and geometric shape. Skeleton extraction is an important step in topological and geometric shape analysis. According to the characteristics of the wheelset laser curve dataset, a new skeleton extraction network, a hierarchical skeleton network (LuoNet), is proposed. The proposed architecture has three levels of the encoder–decoder network, and YE Module interconnection is designed between each level of the encoder and decoder network. In the wheelset laser curve dataset, the F1_score can reach 0.714. Compared with the traditional laser curve center extraction algorithm, the proposed LuoNet algorithm has the advantages of short running time, high accuracy, and stable extraction results.https://www.mdpi.com/1424-8220/22/3/859deep learningsemantic segmentationlaser curve extractionimage processing |
| spellingShingle | Shuai Luo Kai Yang Lijuan Yang Yong Wang Xiaorong Gao Tianci Jiang Chunjiang Li Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network Sensors deep learning semantic segmentation laser curve extraction image processing |
| title | Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network |
| title_full | Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network |
| title_fullStr | Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network |
| title_full_unstemmed | Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network |
| title_short | Laser Curve Extraction of Wheelset Based on Deep Learning Skeleton Extraction Network |
| title_sort | laser curve extraction of wheelset based on deep learning skeleton extraction network |
| topic | deep learning semantic segmentation laser curve extraction image processing |
| url | https://www.mdpi.com/1424-8220/22/3/859 |
| work_keys_str_mv | AT shuailuo lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork AT kaiyang lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork AT lijuanyang lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork AT yongwang lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork AT xiaoronggao lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork AT tiancijiang lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork AT chunjiangli lasercurveextractionofwheelsetbasedondeeplearningskeletonextractionnetwork |