Image processing system for documentary data

Apparatus for extracting information from a document bearing first indicia detectable in a first scan mode using light in a first spectral range but not detectable in a second scan mode using light outside the first spectral range and bearing second indicia detectable in either said first scan mode or second scan mode, said second indicia representing information encoded in a plurality of encoded forms, each encoded form decodable into alphanumeric data by one of plurality of automated symbol recognition resources, comprising:

• (a) means for defining at least two areas of interest on said document, each comprising less than the entire document, in which said second indicia representing information encoded in a plurality of forms may appear, said definition for each area of interest including information specifying the form of encoding for said second indicia and a content test corresponding to the form of encoding to determine if any second indicia in the specified form of encoding are present in the area of interest;
• (b) means for scanning said document in said second mode, using a scanner that produces a first pixel image of the second indicia detected, said first pixel image excluding said first indicia;
• (c) means for storing the pixel data from the first pixel image in a memory device;
• (d) means for selecting one or more areas of interest from those defined for said document, applying the content test defined for the area of interest to said first pixel image and, if the content test is satisfied, processing the first pixel image of said area of interest with one of said plurality of automated symbol recognition resources suitable for the form of encoding specified for the second indicia in the selected area of interest;
• and
• (e) means for producing one or more output records incorporating alphanumeric information representative of the second indicia in said plurality of areas of interest produced by said selecting and processing means.

A method for extracting information from a form document bearing pre-printed indicia and variable indicia responsive to said pre-printed indicia, said variable indicia representing information encoded in a plurality of encoded forms, each encoded form decodable by one of a plurality of automated symbol recognition resources, said method comprising:

• (a) defining at least two areas of interest on said document, each comprising less than the entire document, in which said variable indicia representing information encoded in a plurality of forms may appear, said definition for each area of interest including information specifying the form of encoding for said variable indicia, and a content test corresponding to the form of encoding to determine if any second indicia in the specified form of encoding are present in the area of interest;
• (b) scanning said document using a scanner that produces a pixel image of both pre-printed and variable indicia;
• (c) storing the pixel data obtained from step (b) in a memory device;
• (d) selecting an area of interest from those defined for said document, applying the content test defined for the area of interest and, if the content test is satisfied, processing the pixel image of said area of interest with one of said plurality of automated symbol recognition resources suitable for the form of encoding specified for the variable indicia in the area of interest;
• and
• (e) producing one or more output records incorporating alphanumeric information representative of said variable indicia in said plurality of areas of interest produced by the processing of step (d).