Rare traffic sign detection with synthetic images and multiple classifiers
Detecting rare traffic signs is important for various applications such as autonomous driving, creation of city maps, and road maintenance, as they can provide useful information regarding the surroundings to aid driving decision-making. In this project, we demonstrate that we can train neural netwo...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/136942 |
| _version_ | 1811676561815896064 |
|---|---|
| author | Loke, Yen Chin |
| author2 | Ang Wei Tech |
| author_facet | Ang Wei Tech Loke, Yen Chin |
| author_sort | Loke, Yen Chin |
| collection | NTU |
| description | Detecting rare traffic signs is important for various applications such as autonomous driving, creation of city maps, and road maintenance, as they can provide useful information regarding the surroundings to aid driving decision-making. In this project, we demonstrate that we can train neural networks for the rare traffic sign detection and classification tasks by making use of synthetically rendered images and data augmentation to overcome the lack of rare traffic sign training samples. We also show that two-stage detectors are advantageous in achieving high recall rates on rare traffic signs over single-stage detectors, and is a promising avenue of research with regards to addressing the rare sign detection problem. |
| first_indexed | 2024-10-01T02:23:26Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/136942 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:23:26Z |
| publishDate | 2020 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1369422023-03-04T19:40:46Z Rare traffic sign detection with synthetic images and multiple classifiers Loke, Yen Chin Ang Wei Tech School of Mechanical and Aerospace Engineering The Robotics Institute, Carnegie Mellon University John M. Dolan WTang@ntu.edu.sg Engineering::Computer science and engineering Detecting rare traffic signs is important for various applications such as autonomous driving, creation of city maps, and road maintenance, as they can provide useful information regarding the surroundings to aid driving decision-making. In this project, we demonstrate that we can train neural networks for the rare traffic sign detection and classification tasks by making use of synthetically rendered images and data augmentation to overcome the lack of rare traffic sign training samples. We also show that two-stage detectors are advantageous in achieving high recall rates on rare traffic signs over single-stage detectors, and is a promising avenue of research with regards to addressing the rare sign detection problem. Bachelor of Engineering (Mechanical Engineering) 2020-02-06T05:04:21Z 2020-02-06T05:04:21Z 2019 Final Year Project (FYP) https://hdl.handle.net/10356/136942 en C081 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering Loke, Yen Chin Rare traffic sign detection with synthetic images and multiple classifiers |
| title | Rare traffic sign detection with synthetic images and multiple classifiers |
| title_full | Rare traffic sign detection with synthetic images and multiple classifiers |
| title_fullStr | Rare traffic sign detection with synthetic images and multiple classifiers |
| title_full_unstemmed | Rare traffic sign detection with synthetic images and multiple classifiers |
| title_short | Rare traffic sign detection with synthetic images and multiple classifiers |
| title_sort | rare traffic sign detection with synthetic images and multiple classifiers |
| topic | Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/136942 |
| work_keys_str_mv | AT lokeyenchin raretrafficsigndetectionwithsyntheticimagesandmultipleclassifiers |