Method for reading document entries and addresses

摘要

A method for reading document entries and addresses. Automatic reading comprises intermediate steps: layout analysis, classifying the segmented areas and interpreting the results in a reader as well as video coding if required. According to the invention, image data and the allocated reading results and intermediate results are stored for a relevant random sample during reading in the case of clearly and automatically read information and/or image data and the reading results and intermediate results which are clearly determined by using video coding results in a second automatic reading process are stored for a relevant random sample when video coding is also used to clearly determine information. The automatic reader is newly adapted by said relevant random sample.