Evalution of the methods for detecting the position of car number plate
-
- Fujiyoshi Hironobu
- Chubu University
-
- Umezaki Taizo
- Chubu University
Bibliographic Information
- Other Title
-
- ナンバープレート位置検出法の評価
Search this article
Description
In this paper,we describe about three methods for detecting the position of car number plate.Their features are a labeling of the smoothed image using gaussian filter,template matching using averaged patterns and some artificial neural networks(NNs)based on the Back Propagation algorithm.First,in the case of gaussian filtering method,the smoothed image is used so that the separation of car body and number plate is not sufficient in the making labels.Therefore,a body color of the car influences to the detection score.The template matching is given the high score because of the edge emphasizing.It does not depend on a body color. Finally,we show that the NNs are able to detect with much higher score.
Journal
-
- IEICE technical report. Pattern recognition and understanding
-
IEICE technical report. Pattern recognition and understanding (93), 45-52, 1993
The Institute of Electronics, Information and Communication Engineers
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1570291227409372416
-
- NII Article ID
- 110003299688
-
- NII Book ID
- AN10013232
-
- Text Lang
- ja
-
- Data Source
-
- CiNii Articles
