| 作成者 |
|
|
|
|
|
|
|
| 本文言語 |
|
| 出版者 |
|
|
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 開始ページ |
|
| 終了ページ |
|
| 会議情報 |
|
| 出版タイプ |
|
| アクセス権 |
|
| Crossref DOI |
|
| 権利関係 |
|
| 権利関係 |
|
| 概要 |
Nowadays, with the rapid development of cities, the traffic volume has also increased. Although Taiwan prohibits motorcycles from going on national highways, traffic accidents of motorcycles straying ...into national highways still occur from time to time. This research uses CCTV images as the research data set to intercept the film into images every second as training data for the research and uses image recognition technology combined with deep learning skills to improve the recognition precision of vehicles. According to the analysis and verification results of the research analysis, the precision of the trained vehicle identification can reach over 95%, and the recall rate can reach over 90%; combined with the prototype of the vehicle warning system, if the results of the research can be applied to the entrance of an expressway ramp, It can be immediately identified violating vehicles which Inadvertently, such as motorcycles can be given and blocked in advance, to avoid accidental intrusion of illegal vehicles on national highways.続きを見る
|
Mendeley出力
