Presentation | 2024-03-11 Doppler radar-based recognition of bicycle motions Ryoya Hayashi, Masao Masugi, Kenshi Saho, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this report, we present the results of an experiment in which a Doppler radar-based method was used to detect four behaviors of bicyclists: passing, stopping, turning right, and wandering. A convolutional neural network (CNN), which is a type of deep learning method, was used for discrimination, and then spectrograms calculated from radar information were input as image data. The classification results using CNN, which learned the characteristics of each motion from the input data, show that it is possible to discriminate motions when riding a bicycle. Preliminary movements immediately before the motion were also measured in the same way, demonstrating the applicability of the method to prediction of bicycle movements. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Doppler radar / bicycle / deep learning / motion prediction |
Paper # | ITS2023-80 |
Date of Issue | 2024-03-04 (ITS) |
Conference Information | |
Committee | ITS / IEE-ITS |
---|---|
Conference Date | 2024/3/11(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | BKC, Ritsumeikan Univ. |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Information Processing for ITS, etc. |
Chair | Yusuke Takatori(Kanagawa Inst. of Tech.) / 高橋 聡(名古屋電機工業) |
Vice Chair | Tetsuya Manabe(Saitama Univ.) / Shintaro Ono(Fukuoka Univ.) |
Secretary | Tetsuya Manabe(Ritsumeikan Univ.) / Shintaro Ono(Gunma Univ.) / (日本大学) |
Assistant | Taishi Swabe(NAIST) / 星野 貴弘(日本大学) |
Paper Information | |
Registration To | Technical Committee on Intelligent Transport Systems Technology / Technical Meeting on Intelligent Transport Systems |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Doppler radar-based recognition of bicycle motions |
Sub Title (in English) | |
Keyword(1) | Doppler radar |
Keyword(2) | bicycle |
Keyword(3) | deep learning |
Keyword(4) | motion prediction |
1st Author's Name | Ryoya Hayashi |
1st Author's Affiliation | Ritsumeikan University(Ritsumeikan Univ.) |
2nd Author's Name | Masao Masugi |
2nd Author's Affiliation | Ritsumeikan University(Ritsumeikan Univ.) |
3rd Author's Name | Kenshi Saho |
3rd Author's Affiliation | Ritsumeikan University(Ritsumeikan Univ.) |
Date | 2024-03-11 |
Paper # | ITS2023-80 |
Volume (vol) | vol.123 |
Number (no) | ITS-419 |
Page | pp.pp.7-10(ITS), |
#Pages | 4 |
Date of Issue | 2024-03-04 (ITS) |