Identification number recognition based on random forests

CORC > 厦门大学 > 信息技术－已发表论文

	Identification number recognition based on random forests
	Shi, Hua ; Xie, Yi ; Lin, Chen ; Li, Cuihua ; Zou, Quan ; Xie Y(谢怡) ; Lin C(林琛) ; Li CH(李翠华) ; Zou Q(邹权)
刊名	http://dx.doi.org/10.4156/ijact.vol4.issue 14.28
	2012
关键词	Algorithms Image recognition
英文摘要	The recognition of identification (ID) numbers in ID card is very important in real applications. In this paper, we have proposed a novel algorithm based on random forests to recognize ID numbers in the first-generation ID card, the second-generation ID card and the driving ID card. Firstly we obtain the region of interest by the minimum cross-entropy in spatial domain and determine the type of ID card according to the position of face. After using vertical projection and horizontal projection, we separate ID numbers into different characters and apply random forests to recognize character patches. Additionally, a supplemental method with the encoding rules of ID numbers is introduced to further improve recognition performance. The experiments executed on a challenging dataset have shown that the proposed algorithm has low complexity in computation and a high accuracy rate in recognizing three types of ID cards. Moreover, this algorithm is robust to the changes of illumination, rotation and image distortion.
语种	英语
出版者	Advanced Institute of Convergence Information Technology
内容类型	期刊论文
源URL	[http://dspace.xmu.edu.cn/handle/2288/92820]
专题	信息技术－已发表论文
推荐引用方式 GB/T 7714	Shi, Hua,Xie, Yi,Lin, Chen,et al. Identification number recognition based on random forests[J]. http://dx.doi.org/10.4156/ijact.vol4.issue 14.28,2012.
APA	Shi, Hua.,Xie, Yi.,Lin, Chen.,Li, Cuihua.,Zou, Quan.,...&邹权.(2012).Identification number recognition based on random forests.http://dx.doi.org/10.4156/ijact.vol4.issue 14.28.
MLA	Shi, Hua,et al."Identification number recognition based on random forests".http://dx.doi.org/10.4156/ijact.vol4.issue 14.28 (2012).