CAJ | 학술논문

针对目前车牌定位方法的不足，提出了一种颜色和边缘信息相结合的车牌定位方法。首先根据HSV模型确定车牌底色区域，然后将其与基于边缘检测方法获取的候选车牌区域进行融合，最后通过连通域分析及伪区域去除得到精确定位的车牌。仿真结果表明，定位准确率超过96％，鲁棒性较好，运算速度较快。
침대목전차패정위방법적불족，제출료일충안색화변연신식상결합적차패정위방법。수선근거HSV모형학정차패저색구역，연후장기여기우변연검측방법획취적후선차패구역진행융합，최후통과련통역분석급위구역거제득도정학정위적차패。방진결과표명，정위준학솔초과96％，로봉성교호，운산속도교쾌。
In view of disadvantage of existing locating algorithms,a plate location method based on the color and edge information is proposed.Firstly,the background region of the plate is found according to HSV model, and then this region is fused with candidate license plate regions which are detected by edge detection and mathematical morphology finally,the precise license plate is obtained through analyzing the connected domain. The experiment results show that the accuracy rate of the method is more than 96%, and it has a good robustness and speed.