Abstract
A variety of car navigation systems have been developed for improving convenience of traffic navigation. Regarding construction of maps for car navigation systems, data collection of road navigation signs is necessary. However, the operation for data collection requires so many human tasks and time-consuming. In this paper, we propose a fast and accurate detection method of car navigation signs from videos. First, regions of road navigation signs candidates are extracted by using a color tracking technique robust for illumination changes. Then, the road navigation signs can be detected by applying a shape evaluation to the extracted regions. In order to support actual operations for making car navigation maps, the system automatically judges a best shot from the sequential road sign images by checking their sharpness and sizes. The performance of the system has been evaluated for demonstrating the effectiveness of the proposed method.
| Original language | English |
|---|---|
| Pages (from-to) | 456-464 |
| Number of pages | 9 |
| Journal | Journal of the Institute of Image Electronics Engineers of Japan |
| Volume | 36 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - 2007 Jan |
| Externally published | Yes |
Keywords
- Automatic Detection
- Best Shot Judgement
- Color Tracking
- Road Navigation Sign
- Shape Evaluation
ASJC Scopus subject areas
- Computer Science (miscellaneous)
- Electrical and Electronic Engineering