A new courtesy amount recognition module of CENPARMI's Check Reading System (CRS) is proposed in this paper. The module consists of 3 main segments: pre-processing, segmentation and recognition, and post-processing. A new feedback-based segmentation algorithm is adopted for the segmentation task. Besides one individual numeral recognizer for numerals from '0' to '9', one convolutional neural network (CNN) recognizer for "00" and "000" numeral strings is also integrated into our module for the recognition task. The experimental results on the Quebec Bell Check database show that the recognition rate of the courtesy amount has improved from 41.2% to 74.3%.
展开▼
