电子财务报销入账凭证的自动录入方法研究

由于目前录入电子财务报销入账凭证时,没有对提取的特征进行降噪和去除冗余处理,存在录入效率、录入精准度和录入完整度低的问题.为此,提出电子财务报销入账凭证自动录入方法,通过Gabor滤波器对待录入的凭证数据进行降噪和去冗处理;采用顺序向前搜索方法在多通道的滤波特征图像中提取凭证数据特征;将J-M距离作为判别因子,实现特征...

Research on Automatic Entry Method of Electronic Financial Reimbursement Entrying Vouchers

When the current method is used to enter the electronic financial reimbursement to entry vouchers,the extracted features are not processed for noise reduction and redundancy removal,and there are problems of low entry efficiency,entry accuracy and entry completeness.To this end,an automatic entry method for electronic financial reimbursement entry vouchers is proposed.The Gabor filter is used to reduce noise and redundant processing of the voucher data to be entered,the sequential forward search method is used to extract the voucher data features from the multi-channel filtered feature image,J-M distance is used as a discriminant factor to optimize the feature space.The extracted features are input into the support vector machine,the classification and automatic entry of electronic financial reimbursement can be completed.The experimental results show that the proposed method has high input efficiency,high input accuracy and high input integrity.