| 基于图像配准的纸币图像识别方法 |
| |
| 引用本文： | 徐晓华,周晓飞,金野,李习伦. 基于图像配准的纸币图像识别方法[J]. 智能计算机与应用, 2013, 0(6): 78-80,84 |
| |
| 作者姓名： | 徐晓华 周晓飞 金野 李习伦 |
| |
| 作者单位： | [1]广东百佳百特实业有限公司,广东中山528400 [2]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001 [3]上海大学通信与信息工程学院,上海200000 |
| |
| 基金项目： | 2011年广东省教育部产学研结合项目（20118090400339）. |
| |
| 摘 要： | 提出了基于图像配准的多光谱纸币图像识别架构，通过对纸币图像的配准来实现相似纸币识别、纸币防伪特征检测。由于纸币图像同时具有显著的结构特性和不稳定的局部特性，采用分级的变换模型对纸币图像配准：对于全局的刚性形变采用基于Harris角点特征的仿射变换算法，对于纸币图像局部非刚性形变采用基于B-样条的FFD模型的非刚性配准算法。实验表明，本模型能很好的完成纸币图像的配准，并在相似纸币识别和红外防伪特征检测方面能达到较高的精确度。
|
| 关 键 词： | 纸币图像 图像配准 Harris角点特征 B-样条FFD模型 |
The Method of Banknote Recognition based on Image Registration |
| |
| XU Xiaohua,ZHOU Xiaofei,＇,JIN Ye,LI Xilun. The Method of Banknote Recognition based on Image Registration[J]. INTELLIGENT COMPUTER AND APPLICATIONS, 2013, 0(6): 78-80,84 |
| |
| Authors: | XU Xiaohua ZHOU Xiaofei ＇ JIN Ye LI Xilun |
| |
| Affiliation: | 1 ( 1 PARKnSHOP Baxter Industrial Co., Ltd. zhongshan Guangdong 528400, China; 2 School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China ; 3 School of Communication and Information Engineering, Shanghai university, Shanghai 200000, China) |
| |
| Abstract: | This paper presents a multi - spectral banknote recognition framework based on image registration , which a- chieves similar banknotes recognition and banknotes anti - counterfeit. Banknote has both special structural characteristic and local instability characteristic . The paper uses a hierarchical transformation model： the global deformation of the ban- knote image is modeled by an affine transformation based on Harris corner feature , while the local image＇ s nonrigid deform- ation is described by an FFD（ free- form deformation model）based on B -spline. The experiment shows the model can a- chieve a high precision on similar banknotes recognition and security features detection. |
| |
| Keywords: | Banknote Image Image Registration Harris Corner Feature Free - Form Deformation Model |
| 本文献已被 维普 等数据库收录！ |
|
