研究了纸币退化问题,提出了一种基于颜色退化序列的图像纸币质量评价方法。该方法根据颜色扩散模型建立反映纸币退化程度的图像模式序列,通过该图像模式序列来寻找被检测纸币图像在相应图像序列中的最佳位置,进而对图像质量(新旧、残缺)进行评价。实验结果表明,该方法根据对纸币的退化程度的分析,不但可以对纸币的质量进行评价,还可以提高对低质量纸币的处理能力,大大降低拒识率。而且该方法不但适合纸币图像,还适合其它与之类似的印刷品的质量评价。
The problem of banknote degradation was studied, and an image-based banknote evaluation method based on color degradation sequence was proposed. The method builds an image pattern sequence according to the color diffusion model to measure the degree of color degradation, detects the best position of the banknote image in the image se- quence using the image pattern sequence, and then evaluates the image quality (old, or new, or incomplete). The experimental results show that the algorithm constructed based on analysis of the degradation extent can not only evaluate the quality of banknotes, but also improve the capacity of handling low-quality of banknotes, notes, greatly reducing the rejection rate. The method is not only suitable for the quality evaluation for paper money images, but also for other similar printed stuff.