A Novel Algorithm of Character Segmentation in Vehicle License Plates
Character segmentation for Iicense plate is a key component of an automatic vehicle license recognition system.A novel adaptive approach of character segmentation for badly degraded images is proposed.First.the non-uniform iIlumination correction, contrast enhancement and binarization are introduced;then,op-bottom edges of characters can be obtained by the Units of the plate horizontal projection respectively;finally,the candidate regions of characters are located according to the plate vertical projection for coarse segmentation. And the character segmentation regions are determined by prior knowledge of license plate.The experimental results based on Chinese Iicense plates shows that the proposed method is fast and accurate.and is tolerant to license plate with deformations,rotations,plate frame,rivet,the space mark,and so on.
binarization character segmentation horizontal projection vertical projection
Huang fan Li zhi-min Zhang jing Wan rui Zhang feng-yang
国际会议
The International Conference Information Computing and Automation(2007国际信息计算与自动化会议)
成都
英文
678-680
2007-12-19（万方平台首次上网日期，不代表论文的发表时间）