基于正态分布和向量范数理论的方法,提出用于判断货币识别传感器的有效性和精确性分析方法,并建立了货币识别的性质模型。该模型通过存储概率函数的参数,有效保留了货币的信息,减少存储空间、提高检索速度。考虑了传感器获取的所有信息,能够消除维数的影响,通过实验数据统计表明该模型对指导货币性质研究及识别方法选取有较为实际的应用价值。
Based on the theories of normal distribution and vector norm, a novel method for evaluating the effectiveness and precision of the sensor used in currency recognition is proposed. The model of currency attribute is built and used to save the information of currency attribute by storing the parameter of probability function, which can decrease the memory space for storing information and improve the searching speed. It takes into account all the information that the sensor gains and reduces the dimension effect. Test data statistics shows that this model has practical application value for the study of currency property and the selection of recognition methods.