A New Approachfor License Plate Recognition without Character Partition
碩士 === 靜宜大學 === 資訊管理學系研究所 === 92 === Although many productions of Vehicle License Plate (V.L.P.) Recognition Systems have been applied to the runway of parking lots, they were not extended to mobile environments due to interference. These interferences cause failures on subsystems of “Localization o...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2004
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/9a72w2 |
| Summary: | 碩士 === 靜宜大學 === 資訊管理學系研究所 === 92 === Although many productions of Vehicle License Plate (V.L.P.) Recognition Systems have been applied to the runway of parking lots, they were not extended to mobile environments due to interference. These interferences cause failures on subsystems of “Localization of Vehicle Plates” and “Character Recognition of Vehicle Plates.” This research concentrates on the improvement for “Character Recognition of Vehicle Plates.”
In order to overcome the interference in environments we need a simple and effective method to preprocess images. Thus a new binarization technique is proposed to achieve the requirement under most environments.
This research especially contributes to a approach without partitioning characters. Based on this approach we may avoid failures from partition. However, we have a challenge of recognition speed. A* algorithm is employed to search optimal path. Experiment results show that the proposed approach has high recognition rate and outstanding efficiency.
|
|---|