The design and implementation of a Chinese financial invoice recognition system

This paper designs and implements a financial invoice recognition system based on the features of the Chinese financial invoice. By using the linear whole block moving method in each vertical segment, a new fast algorithm is put forth to detect and rectify the slant image. To distinguish the different form types (the foundation necessary for locating the form fields, filtering the form lines, etc.), several representative form features are discussed and an invoice-type features library is built by using a semi-automatic machine study method. On the basis of the recognized invoice type, real invoice form is re-oriented against the corresponding blank form according to the invoice type feature, solving the problem of adhesion of characters and form lines, as well as the problem of characters segmentation and recognition. Based on the financial Chinese invoice image feature, a mutual rectification mechanism founded on the recognition results of financial Chinese characters and Arabic numerals is put forward to raise the recognition rate. Finally, the experimental results and conclusions are presented.