基于先验知识的香烟防伪数字串识别

DIGITAL IDENTIFICATION FOR CIGARETTE ANTI-COUNTERFEITING BASED ON PRIOR KNOWLEDGE

香烟防伪数字串对于随机抽取市场香烟防伪数字并批量鉴别香烟真伪十分重要。针对香烟防伪鉴定实际需求,提出一种新的数字串识别方法。采用目标检测算法,定位出大概数字串区域的最小外接矩形;将最小外接矩形分割,将分割得到的小矩形图片分类,判断分类结果是否满足先验知识。如果不满足,则调整外接矩形参数,重复以上分割识别过程,直到找到满足要求的矩形,并把此时的分类结果作为最终识别结果。实验结果表明,该方法较好地满足了香烟鉴定的需求。

Cigarette anti-counterfeit numbers are vital for randomly extracting cigarette anti-counterfeiting numbers from the market and batch identification of the authenticity of cigarettes.According to the actual demands of anti-counterfeiting identification of cigarettes,we propose a new method for it.We adopted the object detection algorithm to seek for the minimum circumscribed rectangle of the approximate digital string area.And the minimum circumscribed rectangle was segmented to get small rectangular images for classification so as to determine whether the result satisfied the prior knowledge or not.If not,we adjusted the parameter of the rectangle,repeated segmentation and recognition until we found a rectangle that met the requirements,and took the classification results as the final recognition results.The experimental results show that the identification method meets the needs of cigarette identification.