Classifier performance

Details Classifier performance Classifier performance

2004-10-26

2008-11-11

Bella, Matthew C. (Department: 2624)

Image analysis

Pattern recognition

Classification

C382S159000, C382S118000, C382S115000

Reexamination Certificate

active

07450766

ABSTRACT:
Methods, machines, and machine readable media storing machine-readable instructions for improving classifier performance are described. In one aspect, threshold vectors Tk=(t1k, t2k, . . . , tnk) are input into a classifier that includes a cascade of n classification stages. Each classification stage i has a respective classification boundary that is controlled by a respective threshold tik, wherein i=1, . . . , n, n has an integer value greater than 1, k is an integer, and each threshold vector Tkcorresponds to a respective operating point of the classifier. Threshold vectors corresponding to respective points on a receiver operating characteristic (ROC) curve approximating a target discrimination performance for multiple operating points of the classifier are identified.

REFERENCES:
patent: 6205247 (2001-03-01), Breuer et al.
patent: 6546379 (2003-04-01), Hong et al.
patent: 7050607 (2006-05-01), Li et al.
patent: 7099510 (2006-08-01), Jones et al.
patent: 2002/0102024 (2002-08-01), Jones et al.
patent: 2002/0110279 (2002-08-01), Dobeck
patent: 2004/0013304 (2004-01-01), Viola et al.
patent: 2004/0066966 (2004-04-01), Schneiderman
patent: 2005/0013479 (2005-01-01), Xiao et al.
patent: 2005/0214760 (2005-09-01), Chan et al.
Yoav Freund et al., “A Short Introduction to Boosting,” Journal of Japanese Society for Artificial Intelligence, 14(5):771-780 (Sep. 1999).
Paul Viola et al., “Rapid Object Detection using a Boosted Cascade of Simple Features,” Conf. on Computer Vision & Pattern Recognition (2001).
Paul Viola et al., “Robust Real-time Object Detection,” Cambridge Research Laboratory, Technical Report Series (Feb. 1999).

Affiliated with

Also associated with

Rating

Classifier performance does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Classifier performance, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Classifier performance will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4039473