Methods and systems for extraction of data from electronic images of documents

摘要

A system for extracting data from a document may include a memory, an interface, and a processor. The processor may identify one or more landmarks based on predefined characteristics that are associated with known landmarks. After one or more landmarks are defined, business rules may indicate one or more areas that may contain data to be extracted. The business rules may also indicate the type of data in the area and processing methods that may be used to extract the data. After determining the business rules, the processor may determine whether the data is in the area and/or extract the data.

主权项

1. A method of extracting data from an electronic image of a document, the document having at least one indicium therein indicative of one or more designated areas, separate from the indicium, in which the data may be found, the method comprising:
(a) identifying each of the at least one indicium; (b) determining, for each identified indicium, at least one rule associated with the identified indicium, wherein each of the determined at least one rule provides information regarding the data to be extracted and designates the one or more areas, separate from the indicium, in which the data to be extracted may be found, and each of the at least one rule further provides, wherein the information regarding the data to be extracted comprises type of data to be extracted, for how to extract the data, threshold for determining if the area contains the data to be extracted, or combinations thereof; and (c) extracting data from only the designated one or more areas of the electronic image, wherein at least a portion of the information provided by the determined at least one rule guides extraction of the data from the one or more areas designated thereby.