Image analysis – Pattern recognition – On-line recognition of handwritten characters
Reexamination Certificate
1999-01-28
2003-10-14
Dastouri, Mehrdad (Department: 2623)
Image analysis
Pattern recognition
On-line recognition of handwritten characters
C215S325000, C215S325000
Reexamination Certificate
active
06633671
ABSTRACT:
FIELD
The present invention teaches a system of using a camera to track handwriting and determine certain characteristics of the handwriting. More specifically, the present system shows a first embodiment with a monocular way of tracking the movement of the pen and also a camera-based way of identifying a person's identity.
BACKGROUND
Advances in computer technology have given rise to smaller computing systems. As the computing structure becomes smaller, the size of the keyboard and the size of the screen become a real limit on the size of the computer. For example, the laptop computer cannot be easily made smaller than the keyboard. One solution is to use data entry via handwriting. Many handwriting systems use a tablet that detects the pressure of the pen. The pressure-sensitive tablets, however, are specialized materials in themselves and take up space.
Security is an important issue in current information technology, and it is highly desirable to be able to identify a person using biometric aspects. Various people have suggested different biometric techniques, including face recognition, fingerprint recognition, iris recognition, and others.
SUMMARY
The present inventors disclose systems which use a video camera to capture the motion of the hand and translate that motion into handwriting. The principles are described in U.S. patent application Ser. No. 081,490,678, now U.S. Pat. No. 6,044,165. This system uses an image acquisition device, e.g. a camera, a writing surface, e.g., a piece of paper, and a writing implement, e.g. a normal pen, as the data entry device. The only necessary hardware that is not common is the camera. However, cameras can be made much smaller than keyboards.
During processing of the system, the inventors found that it is advantageous to determine when the pen is up and when the pen is down. One aspect of the present system describes a way of tracking when the pen is up and down using a monocular system.
Another aspect uses both the pen up and pen down strokes to form an identification technique using biometric techniques. The present system also describes a special way of verifying identity using signature verification and special features in the signature verification.
REFERENCES:
patent: 5077802 (1991-12-01), Plamondon
patent: 5103486 (1992-04-01), Grippi
patent: 5226091 (1993-07-01), Howell et al.
patent: 5454046 (1995-09-01), Carman, II
patent: 5577135 (1996-11-01), Grajski et al.
patent: 5617312 (1997-04-01), Iura et al.
patent: 5661506 (1997-08-01), Lazzouni et al.
patent: 5745592 (1998-04-01), Nalwa
patent: 5771036 (1998-06-01), Martin et al.
patent: 5802200 (1998-09-01), Wirtz
patent: 5878164 (1999-03-01), Brown et al.
patent: 6044165 (2000-03-01), Perona et al.
patent: 6052481 (2000-04-01), Grajski et al.
patent: 6249604 (2001-06-01), Huttenlocher et al.
Munich et al, Visual Input for Pen-Based Computers, Aug. 1996, IEEE ISBN: 0-8186-7282-X, vol. 3, pp. 33-37.*
Vishvjit S. Nalwa; Automatic On-Line Signature Verification; Feb. 1997; Proceedings of the IEEE, vol. 85. No. 2. pp. 215-239.
Munich Mario E.
Perona Pietro
California Institute of Technology
Dastouri Mehrdad
Fish & Richardson P.C.
LandOfFree
Camera-based handwriting tracking does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Camera-based handwriting tracking, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Camera-based handwriting tracking will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3145186