Device and method for verifying personal handwriting

• US 5,095,510 A
• Filed: 01/07/1991
• Issued: 03/10/1992
• Est. Priority Date: 02/16/1990
• Status: Expired due to Term

• Current Assignee: Rolls-Royce PLC (Rolls-Royce Holdings Plc)
• Original Assignee: Rolls-Royce PLC (Rolls-Royce Holdings Plc)
• Inventors: Webster, John R.
• Primary Examiner(s): Mancuso, Joseph
• Application Number:
  US07/637,704
• Time in Patent Office: 428 Days
• Field of Search: 382/3, 382/34, 382/13, 382/30, 178/18, 178/19, 178/20
• US Class Current: 382/121
• CPC Class Codes: G06F 18/28 Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
• CPC Class Codes: G06V 40/30 Writer recognition; Reading and verifying signatures
• CPC Class Codes: G06V 40/37 based only on signature signals such as velocity or pressure, e.g. dynamic signature recognition
• CPC Class Codes: G07C 9/35 by means of a handwritten signature

Abstract

The stress wave activity produced in a surface by a writing instrument during the generation of a given piece of handwriting is detected by a transducer and is electrically processed. The stress wave activity is compared with a stress wave activity, produced during at least one previous generation of the given piece of handwriting, held in a store, by a processing unit in order to determine whether the detected and stored handwriting pieces have a common author. If the detected and stored handwriting pieces are determined to have a common author a weighted average of the stored stress wave activity and the detected stress wave activity is produced. The weighted average includes a weighted average of the amplitudes and a weighted average of times of events. This weighted average stress wave activity then replaces the stored stress wave activity in the store. This allows natural variations in the way an individual writes a given piece of handwriting to be accommodated without false indications of inauthenticity.

Claims

1. A device for verifying personal handwriting including a first body on which a piece of handwriting is written using a first writing instrument, said device comprising:

means for detecting a time varying characteristic produced by the motion of said writing instrument on a surface of said body during the generation of said piece of handwriting,

storage means for storing a time varying characteristic produced by the motion of a second writing instrument similar to said first writing instrument on a surface of a second body similar to said first body during the generation of at least one previous like piece of handwriting,

means for comparing the detected time varying characteristic with the stored time varying characteristic to provide an indication of whether the generation of said handwriting pieces produce substantially common time varying characteristics and thus whether the handwriting pieces have been generated by a common author, the means for comparing the detected time varying characteristic with the stored time varying characteristic being arranged to compare each point of the detected time varying characteristic with a corresponding point of the stored time varying characteristic,

means for producing a weighted average of the stored time varying characteristic and the detected time varying characteristic produced during the generation of a piece of handwriting by a common author, the producing means producing a weighted average of magnitudes of the stored time varying characteristic and the detected time varying characteristic as well as producing a weighted average of time of events of the stored time varying characteristic and the detected time varying characteristic, and

means for replacing the stored time varying characteristic with the weighted average of the stored time varying characteristic and the detected time varying characteristic produced during the generation of a piece of handwriting by a common author.

2. A device as claimed in claim 1 in which the means for comparing the detected time varying characteristic with the stored time varying characteristic is arranged to distort each point of said detected time varying characteristic in time to obtain a best match with the stored time varying characteristic and stores each of the values of time distortion required to obtain the best match.

3. A device as claimed in claim 2 in which the means for producing a weighted average of the detected time varying characteristic and the stored time varying characteristic is arranged to distort each point of the detected time varying characteristic in time by the corresponding one of the values stored by the comparing means to obtain the best match with the stored time varying characteristics, and is arranged to produce weighted average magnitudes of said time varying characteristic between the stored time varying characteristic and the time distorted detected time varying characteristic and is arranged to produce a weighted average of time of events by displacing the weighted average magnitudes of said time varying characteristic in time towards the detected time varying characteristic.

4. A device as claimed in claim 2 in which the means for comparing the detected time varying characteristic with the stored time varying characteristic is arranged to detect relatively low or zero magnitudes of the time varying characteristic and is arranged to reduce the errors associated with the stored values of time distortion required to obtain the best match with the stored time varying characteristic.

5. A device as claimed in claim 1 in which the time varying characteristic is capacitance, the means for detecting the time varying characteristic is a capacitor defined between the writing instrument and the surface of the body.

6. A device as claimed in claim 1 in which the time varying characteristic is load, the means for detecting the time varying characteristic is a load cell for detecting load produced by the motion of the writing instrument on the surface of the body.

7. A device as claimed in claim 1 in which the time varying characteristic is stress wave activity, the means for detecting the time varying characteristic is an acoustic emission transducer for detecting the stress wave activity produced by the motion of the writing instrument on the surface of the body.

8. A device as claimed in claim 7 in which the means for detecting the stress wave activity comprises a resonant acoustic emission transducer.

9. A device as claimed in claim 7 comprising a means for processing the detected stress wave activity including an enveloping means.

10. A device for verifying personal handwriting comprising the steps of:

detecting a time varying characteristic produced by the motion of a writing instrument on a surface of a body during the generation of a given piece of handwriting,

comparing the detected time varying characteristic with a stored time varying characteristic produced by the motion of a like writing instrument on a like surface of a like body during the generation of at least one previous piece of handwriting to provide an indication of whether the generation of said handwriting pieces produce common time varying characteristics and thus whether the pieces have been generated by a common author,

comparing each point of the detected time varying characteristic with a corresponding point of the stored time varying characteristic,

producing a weighted average of the stored time varying characteristic and the detected time varying characteristic produced during the generation of said handwriting by a common author, the producing step producing a weighted average of magnitudes of the stored time varying characteristic and the detected time varying characteristic and a weighted average of time of events of the stored time varying characteristic and the detected time varying characteristic,

replacing the stored time varying characteristic with the weighted average of the stored time varying characteristic and the detected time varying characteristic produced during the generation of a piece of handwriting by a common author.

11. A method as claimed in claim 10 in which each point of the detected time varying characteristic is distorted in time to obtain a best match with the stored time varying characteristic, each of the values of time distortion required to obtain the best match is stored.

12. A method as claimed in claim 11 in which each point of the detected time varying characteristic is distorted in time by the corresponding one of the stored values to obtain the best match with the stored time varying characteristic, producing weighted average magnitudes of said time varying characteristic between the stored time varying characteristic and the time distorted detected time varying characteristic and producing a weighted average of time of events by displacing the weighted average magnitudes of said time varying characteristic in time towards the detected time varying characteristic.

13. A method as claimed in claim 10 in which the time varying characteristic detected is capacitance.

14. A method as claimed in claim 10 in which the time varying characteristic detected is load or pressure.

15. A method as claimed in claim 10 in which the time varying characteristic detected is stress wave activity.