Database entry system and method employing optical character recognition

摘要

A system is disclosed for extracting data from non-standardized paper formats into electronic files, such as databases. The system is particularly suited for compiling data from engineering drawings and paper catalogs. A user interface presents an image (1114) of the paper document along with a list of attributes (1212) from a database. The user identifies regions of the image that contain relevant data, and highlights those regions on the computer screen (1316). The system incorporates optical character recognition (1410) to convert images of alphanumeric characters within each highlighted region into computer readable data. The user associates each highlighted region with an attribute (1318) so that data from the region gets stored as an attribute value. In one embodiment of the invention, the user interface is constructed to process columns of data directly into spreadsheets. One benefit of the invention is that parts lists can be quickly compiled and edited from engineering drawings. Another benefit is that information from paper catalogs can be quickly converted into computer data and subsequently stored into electronic databases.