Unfolded convolution for fast feature extraction

Details Unfolded convolution for fast feature extraction Unfolded convolution for fast feature extraction

2005-10-14

2009-12-15

Ahmed, Samir A. (Department: 2624)

Image analysis

Pattern recognition

Feature extraction

C382S279000, C382S132000, C382S128000

Reexamination Certificate

active

07634137

ABSTRACT:
Systems and methods are described that facilitate performing feature extraction across multiple received input features to reduce computational overhead associated with feature processing related to, for instance, optical character recognition. Input feature information can be unfolded and concatenated to generate an aggregated input matrix, which can be convolved with a kernel matrix to produce output feature information for multiple output features concurrently.

REFERENCES:
patent: 5577135 (1996-11-01), Grajski et al.
patent: 6408107 (2002-06-01), Miller et al.
patent: 6483940 (2002-11-01), Wang
patent: 6608929 (2003-08-01), Shiratani
patent: 6947502 (2005-09-01), Taylor et al.
patent: 7231080 (2007-06-01), Hakim et al.
Patrice Y.Simard, Leon Bottou, Patrick Haffner, Yann Le Cun, Boxlets: a fast convolution Algorithm for Signal processing and Neural Networks(pdf), In Advance in Neural Information Processings Systems, Kearns, M., Solla, S., Cohn, D. (Eds.), vol. 11, MIT Press, pp. 571-577, 11999.

Affiliated with

Also associated with

Rating

Unfolded convolution for fast feature extraction does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Unfolded convolution for fast feature extraction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Unfolded convolution for fast feature extraction will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4125064