Accession Number : AD0803633

Title :   IMPLEMENTATION OF DOCUMENT FORMAT RECOGNITION,

Descriptive Note : Final rept. May 65-Sep 66,

Corporate Author : SYLVANIA ELECTRONIC SYSTEMS-EAST WALTHAM MA APPLIED RESEARCH LAB

Personal Author(s) : Blitz, M. ; Stoddard, J. ; Sanders, R. ; Robinson, M. B. ; Sallen, R.

Report Date : OCT 1966

Pagination or Media Count : 222

Abstract : The design and construction of an experimental model of an automatic system for the recognition of the page format of foreign and domestic journals are described. The system consisted of a laboratory page reader, a PDP-8 control computer, and a CDC 3200 general purpose computer. Operationally, such a system would be used in information retrieval and language translation applications. During the initial phases, the format recognition and analysis programs were designed and coded under the assumption that the page reader would be able to recognize characters on the journal page. When it became apparent that the flying spot scanner in the laboratory page reader had insufficient resolution for the simple separation of characters, these programs were revised to permit an investigation of the feasibility of document format recognition without character recognition. While it was possible to perform rudimentary format recognition such as distinguishing between text and graphics, the system performance was far below that possible with character recognition. It is recommended that further work in this area be directed toward the improvement of the page reader.

Descriptors :   *READING MACHINES), (*PATTERN RECOGNITION, (*CHARACTER RECOGNITION, AUTOMATIC), REPORTS, INFORMATION RETRIEVAL, MACHINE TRANSLATION, REAL TIME, COMPUTER PROGRAMMING, FLOW CHARTING, OPTICAL SCANNING.

Subject Categories : Information Science
      Computer Programming and Software

Distribution Statement : APPROVED FOR PUBLIC RELEASE