Determining bus stop locations using deep learning and time filtering
This paper presents an intelligent bus stop determination from bus Global Positioning System (GPS) trajectories. A mixture of deep neural networks and a time filtering algorithm is used in the proposed algorithm. A deep neural network uses the speed histogram and azimuth angle at each location as in...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/161636 |
| _version_ | 1824454360414814208 |
|---|---|
| author | Piriyataravet, Jitpinun Kumwilaisak, Wuttipong Chinrungrueng, Jatuporn Piriyatharawet, Teerawat |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Piriyataravet, Jitpinun Kumwilaisak, Wuttipong Chinrungrueng, Jatuporn Piriyatharawet, Teerawat |
| author_sort | Piriyataravet, Jitpinun |
| collection | NTU |
| description | This paper presents an intelligent bus stop determination from bus Global Positioning System (GPS) trajectories. A mixture of deep neural networks and a time filtering algorithm is used in the proposed algorithm. A deep neural network uses the speed histogram and azimuth angle at each location as input features. A deep neural networks consists of the convolutional neural networks (CNN), fully connected networks, and bidirectional Long-Short Term Memory (LSTM) networks. It predicts the soft decisions of bus stops at all locations along the route. The time filtering technique was adopted to refine the results obtained from the LSTM net-work. The time histograms of all locations was built where the high potential timestamps are extracted. Then, a linear regression is used to produce an approximate reliable timestamp. Each time distribution can be derived using data updated at that time slot and compared to a reference distribution. Locations are predicted as bus stop locations when timestamp distributions close to the reference distributions. Our technique was tested on real bus service GPS data from National Science and Technology Development Agency (NATDA, Thailand). The proposed method can outperform other existing bus stop detection systems. |
| first_indexed | 2025-02-19T03:21:04Z |
| format | Journal Article |
| id | ntu-10356/161636 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:21:04Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | ntu-10356/1616362022-09-13T02:11:07Z Determining bus stop locations using deep learning and time filtering Piriyataravet, Jitpinun Kumwilaisak, Wuttipong Chinrungrueng, Jatuporn Piriyatharawet, Teerawat School of Computer Science and Engineering Engineering::Computer science and engineering Global Positioning System Convolutional Neural Network This paper presents an intelligent bus stop determination from bus Global Positioning System (GPS) trajectories. A mixture of deep neural networks and a time filtering algorithm is used in the proposed algorithm. A deep neural network uses the speed histogram and azimuth angle at each location as input features. A deep neural networks consists of the convolutional neural networks (CNN), fully connected networks, and bidirectional Long-Short Term Memory (LSTM) networks. It predicts the soft decisions of bus stops at all locations along the route. The time filtering technique was adopted to refine the results obtained from the LSTM net-work. The time histograms of all locations was built where the high potential timestamps are extracted. Then, a linear regression is used to produce an approximate reliable timestamp. Each time distribution can be derived using data updated at that time slot and compared to a reference distribution. Locations are predicted as bus stop locations when timestamp distributions close to the reference distributions. Our technique was tested on real bus service GPS data from National Science and Technology Development Agency (NATDA, Thailand). The proposed method can outperform other existing bus stop detection systems. Published version 2022-09-13T02:11:07Z 2022-09-13T02:11:07Z 2021 Journal Article Piriyataravet, J., Kumwilaisak, W., Chinrungrueng, J. & Piriyatharawet, T. (2021). Determining bus stop locations using deep learning and time filtering. Engineering Journal, 25(8), 163-172. https://dx.doi.org/10.4186/ej.2021.25.8.163 0125-8281 https://hdl.handle.net/10356/161636 10.4186/ej.2021.25.8.163 2-s2.0-85114479918 8 25 163 172 en Engineering Journal © 2021 Faculty of Engineering, Chulalongkorn University, Phayathai Road, Pathumwan, Bangkok 10330, Thailand. This is an open-access article distributed under the terms of the Creative Commons Attribution License. application/pdf |
| spellingShingle | Engineering::Computer science and engineering Global Positioning System Convolutional Neural Network Piriyataravet, Jitpinun Kumwilaisak, Wuttipong Chinrungrueng, Jatuporn Piriyatharawet, Teerawat Determining bus stop locations using deep learning and time filtering |
| title | Determining bus stop locations using deep learning and time filtering |
| title_full | Determining bus stop locations using deep learning and time filtering |
| title_fullStr | Determining bus stop locations using deep learning and time filtering |
| title_full_unstemmed | Determining bus stop locations using deep learning and time filtering |
| title_short | Determining bus stop locations using deep learning and time filtering |
| title_sort | determining bus stop locations using deep learning and time filtering |
| topic | Engineering::Computer science and engineering Global Positioning System Convolutional Neural Network |
| url | https://hdl.handle.net/10356/161636 |
| work_keys_str_mv | AT piriyataravetjitpinun determiningbusstoplocationsusingdeeplearningandtimefiltering AT kumwilaisakwuttipong determiningbusstoplocationsusingdeeplearningandtimefiltering AT chinrungruengjatuporn determiningbusstoplocationsusingdeeplearningandtimefiltering AT piriyatharawetteerawat determiningbusstoplocationsusingdeeplearningandtimefiltering |