Browse Prior Art Database

Optical Character Recognition with Programmable Logic Arrays

IP.com Disclosure Number: IPCOM000084291D
Original Publication Date: 1975-Oct-01
Included in the Prior Art Database: 2005-Mar-02
Document File: 3 page(s) / 48K

Publishing Venue

IBM

Related People

Brickman, NF: AUTHOR

Abstract

An implementation of the recognition logic of an Optical Character Recognition (OCR) machine using Programmable Logic Arrays (PLAs) is presented.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Optical Character Recognition with Programmable Logic Arrays

An implementation of the recognition logic of an Optical Character Recognition (OCR) machine using Programmable Logic Arrays (PLAs) is presented.

The basic OCR organization is shown in Fig. 1. The recognition logic or RECO logic 10 takes the optical scan data from an optical wand 12 and operates on it with Boolean logic expressions, to recognize the presence of characters in the video field. The optical wand 12 is organized as shown in Fig. 2 with 40 rows and 12 columns. The wand 12 sends its data from the bottom row up, one row (12 bits) in parallel at a time to the RECO logic 10, with character recognition being based upon a field of 16 contiguous rows and all 12 columns.

The organization of PLAs to perform the recognition logic is shown in Fig. 3. Each box shown is a different PLA (chip). The so-called first level of PLA chips has a prefix of 1 before the delta or triangle sign, and similarly the second level of chips has a prefix of 2. The PLAs have external inputs and external outputs, as shown in Fig. 3, but also have an internal feedback with JK latches in the feedback path to give storage capability.

Each row of 12 bits from the optical wand 12 goes first to chip C1 delta 2. Each chip in level 1 stores one row of data from the wand. When a new row of data from the wand is available, the old data is shifted over by one chip; for example, the row stored on C1 delta 2 goes to C1 delta 3, C1 delta 3 goes to C1 delta 4, etc. The newest row is stored in latches external to the inputs of chip C1 delta 2. If big enough PLAs are available, more than one row of the wand can be stored on a PLA and the chip count decreases, and the logical recognition operations can be simplified as well.

Several PLA...