Speeding up the generalized adaptive neural filters

Henry Hanek; Nirwan Ansari

doi:10.1109/83.495954

Speeding up the generalized adaptive neural filters

Henry Hanek, Nirwan Ansari

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

13 Scopus citations

Abstract

Recently, a new class of adaptive filters called generalized adaptive neural filters (GANF's) emerged. They share many things in common with stack filters and include all stack filters as a subset. The GANF's allow a very efficient hardware implementation once they are trained. However, the training process can be slow. This paper discusses structural modifications to allow for faster training. In addition, these modifications can lead to an increase in the filter's robustness, given a limited amount of training data. This paper does not attempt to justify use of a GANF; it only presents an alternative implementation of the filter.

Original language	English (US)
Pages (from-to)	705-712
Number of pages	8
Journal	IEEE Transactions on Image Processing
Volume	5
Issue number	5
DOIs	https://doi.org/10.1109/83.495954
State	Published - 1996

All Science Journal Classification (ASJC) codes

Software
Computer Graphics and Computer-Aided Design

Access to Document

10.1109/83.495954

Cite this

@article{b93e02ce05824c5292504e73baca600a,

title = "Speeding up the generalized adaptive neural filters",

abstract = "Recently, a new class of adaptive filters called generalized adaptive neural filters (GANF's) emerged. They share many things in common with stack filters and include all stack filters as a subset. The GANF's allow a very efficient hardware implementation once they are trained. However, the training process can be slow. This paper discusses structural modifications to allow for faster training. In addition, these modifications can lead to an increase in the filter's robustness, given a limited amount of training data. This paper does not attempt to justify use of a GANF; it only presents an alternative implementation of the filter.",

author = "Henry Hanek and Nirwan Ansari",

note = "Funding Information: Manuscript received January 30, 1994; revised August 31, 1995. This work was supported in part by the New Jersey Commission on Science and Technology under Contract IP96-2890-040-18, The associate editor coordinating the review of this paper and approving it for publication was Prof. Dan Schonfeld.",

year = "1996",

doi = "10.1109/83.495954",

language = "English (US)",

volume = "5",

pages = "705--712",

journal = "IEEE Transactions on Image Processing",

issn = "1057-7149",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "5",

}

TY - JOUR

T1 - Speeding up the generalized adaptive neural filters

AU - Hanek, Henry

AU - Ansari, Nirwan

N1 - Funding Information: Manuscript received January 30, 1994; revised August 31, 1995. This work was supported in part by the New Jersey Commission on Science and Technology under Contract IP96-2890-040-18, The associate editor coordinating the review of this paper and approving it for publication was Prof. Dan Schonfeld.

PY - 1996

Y1 - 1996

N2 - Recently, a new class of adaptive filters called generalized adaptive neural filters (GANF's) emerged. They share many things in common with stack filters and include all stack filters as a subset. The GANF's allow a very efficient hardware implementation once they are trained. However, the training process can be slow. This paper discusses structural modifications to allow for faster training. In addition, these modifications can lead to an increase in the filter's robustness, given a limited amount of training data. This paper does not attempt to justify use of a GANF; it only presents an alternative implementation of the filter.

AB - Recently, a new class of adaptive filters called generalized adaptive neural filters (GANF's) emerged. They share many things in common with stack filters and include all stack filters as a subset. The GANF's allow a very efficient hardware implementation once they are trained. However, the training process can be slow. This paper discusses structural modifications to allow for faster training. In addition, these modifications can lead to an increase in the filter's robustness, given a limited amount of training data. This paper does not attempt to justify use of a GANF; it only presents an alternative implementation of the filter.

UR - http://www.scopus.com/inward/record.url?scp=0030151627&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030151627&partnerID=8YFLogxK

U2 - 10.1109/83.495954

DO - 10.1109/83.495954

M3 - Article

C2 - 18285160

AN - SCOPUS:0030151627

SN - 1057-7149

VL - 5

SP - 705

EP - 712

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

IS - 5

ER -

Speeding up the generalized adaptive neural filters

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this