Headlight recognition for night-time traffic surveillance using spatial–temporal information
| Autor: | Toshiaki Kondo, Atsuo Yoshitaka, Sorn Sooksatra, Pished Bunnun |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Similarity (geometry)
Computer science business.industry 020206 networking & telecommunications 02 engineering and technology Motion (physics) Task (project management) Signal Processing 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Multimedia information systems Computer vision Artificial intelligence Electrical and Electronic Engineering F1 score business Temporal information |
| Zdroj: | Signal, Image and Video Processing. 14:107-114 |
| ISSN: | 1863-1711 1863-1703 |
| DOI: | 10.1007/s11760-019-01530-4 |
| Popis: | Vehicle headlights are the important objects especially in the application of night-time traffic surveillance. A common problem of this task is the similarity between the headlights and their reflections on the road. This paper proposes a novel algorithm to construct 3D motion trajectories of headlights and their reflections on the road using both spatial and temporal information. 3D structure tensors are utilized as shape features for recognizing the headlights in various traffic views. Experimental results show that the proposed method performs better than traditional approaches (about 10 $$\%$$) in terms of the F1 score. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink