| 摘要 |
<p>A data processing system and method manages the recognition of text characters in a plurality of document images. The documents are predefined forms having a plurality of fields in which text characters are to be recognized. A form definition data set is used to specify various characteristics of the fields on the form such as their locations. A data capture priority value is specified in the form definition data set for at least some of the fields. This priority is used to manage the workload of the character recognition process. When a large number of document form images have been received by the system, they can limit the backlog volume and maintain a desired throughput for character recognizing the forms by reducing the number of fields which are character recognized on each form. By using the data capture priority value for each field, the character recognition processor can be controlled to omit processing or perform faster, lower-quality processing of low priority fields. <IMAGE></p> |
