Browse Prior Art Database

Thresholding Method for a Mark Reading System

IP.com Disclosure Number: IPCOM000078550D
Original Publication Date: 1973-Jan-01
Included in the Prior Art Database: 2005-Feb-26
Document File: 3 page(s) / 53K

Publishing Venue

IBM

Related People

Dorr, RL: AUTHOR [+3]

Abstract

This thresholding system describes a method for automatically and dynamically altering the thresholding function of a Mark Read Scanner. A General Purpose Mark Reading System must read hand-marked documents from a wide range of applications Typically, marks read from various applications do not have the same density from application-to-application. Valid marks of one application can be equal in density or sensing level to that of normal erasures of another application.

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

Thresholding Method for a Mark Reading System

This thresholding system describes a method for automatically and dynamically altering the thresholding function of a Mark Read Scanner. A General Purpose Mark Reading System must read hand-marked documents from a wide range of applications Typically, marks read from various applications do not have the same density from application-to-application. Valid marks of one application can be equal in density or sensing level to that of normal erasures of another application.

As each row within a field is read, clock 1 cycles threshold levels L(i) (L(i), L(2)...L(N)) to obtain the optimum threshold for Scanner 2 to read that particular data. The "blackest" mark, if any, is found to lie within one of N ranges (delta T(1) delta T(2)... delta T(N)) by the Range decision Logic 3. The optimum threshold range is then recorded in a counter in the counter matrix 4.

Statistical data is gathered over the length of the document by accumulating the delta T ranges in their respective counters. Each counter is reset at the start of read. At the end of the document, discriminator 5 determines which counter has maximum count. The output from the discriminator then gates the proper record from memory to the output device, as shown in Fig. 1.

To utilize contextual information consisting of adjacent marks and/or erasures within a row of the field, levels are combined to provide discrimination. The levels are combined two at a time L(i) and L(i) where L(i) > L(j). The discrimination is determined as follows: (1) Multiple channels within the row having signals above L(i). (2) One signal only above L(i). (3) One signal only above L(j) but not above L(i). (4) No signals above L(j). Case (1) is always a reject. Case (4) also results in a reject condition or a blank condition, as defined by external formatting. Cases (2) and (3) are acceptable ma...