Texture recognition using a superfast cellular neural network VLSI chip in a real experimental environment

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We have developed a new single-chip texture classifier smart-sensor system. Its main part is a programmable cellular neural network (CNN) VLSI chip with optical input and an execution time of a few microseconds. This chip contains a dynamic and locally interconnected 2D cell array. It executes a theoretically new method for texture classification, compared to the other convolution-based feature extraction methods, since here we have feedback convolution as well. Depending on the kernel parameters, this array can execute filtering, moving, linear and nonlinear effects at the same time. The parameters of the feedback and feed-forward convolutions are optimized through a genetic learning using the 22 * 20 CNN chip itself. This chip has a simplified architecture with binary input/output, but it gives good recognition results: this CNN chip with a simple 3 * 3 kernel can reliably classify 5 natural textures. Using more templates for decision-making, more textures can be separated and a classification error of less than 1% has been achieved.

Original languageEnglish
Pages (from-to)1329-1334
Number of pages6
JournalPattern Recognition Letters
Volume18
Issue number11-13
Publication statusPublished - Nov 1997

Fingerprint

Cellular neural networks
Textures
Convolution
Feedback
Smart sensors
Feature extraction
Classifiers
Decision making

Keywords

  • Cellular neural network
  • Deconvolution
  • Genetic algorithm
  • Smart sensors
  • Texture analysis

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Texture recognition using a superfast cellular neural network VLSI chip in a real experimental environment. / Szirányi, T.

In: Pattern Recognition Letters, Vol. 18, No. 11-13, 11.1997, p. 1329-1334.

Research output: Contribution to journalArticle

@article{c11a12c74dd0492e895136064812983d,
title = "Texture recognition using a superfast cellular neural network VLSI chip in a real experimental environment",
abstract = "We have developed a new single-chip texture classifier smart-sensor system. Its main part is a programmable cellular neural network (CNN) VLSI chip with optical input and an execution time of a few microseconds. This chip contains a dynamic and locally interconnected 2D cell array. It executes a theoretically new method for texture classification, compared to the other convolution-based feature extraction methods, since here we have feedback convolution as well. Depending on the kernel parameters, this array can execute filtering, moving, linear and nonlinear effects at the same time. The parameters of the feedback and feed-forward convolutions are optimized through a genetic learning using the 22 * 20 CNN chip itself. This chip has a simplified architecture with binary input/output, but it gives good recognition results: this CNN chip with a simple 3 * 3 kernel can reliably classify 5 natural textures. Using more templates for decision-making, more textures can be separated and a classification error of less than 1{\%} has been achieved.",
keywords = "Cellular neural network, Deconvolution, Genetic algorithm, Smart sensors, Texture analysis",
author = "T. Szir{\'a}nyi",
year = "1997",
month = "11",
language = "English",
volume = "18",
pages = "1329--1334",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",
number = "11-13",

}

TY - JOUR

T1 - Texture recognition using a superfast cellular neural network VLSI chip in a real experimental environment

AU - Szirányi, T.

PY - 1997/11

Y1 - 1997/11

N2 - We have developed a new single-chip texture classifier smart-sensor system. Its main part is a programmable cellular neural network (CNN) VLSI chip with optical input and an execution time of a few microseconds. This chip contains a dynamic and locally interconnected 2D cell array. It executes a theoretically new method for texture classification, compared to the other convolution-based feature extraction methods, since here we have feedback convolution as well. Depending on the kernel parameters, this array can execute filtering, moving, linear and nonlinear effects at the same time. The parameters of the feedback and feed-forward convolutions are optimized through a genetic learning using the 22 * 20 CNN chip itself. This chip has a simplified architecture with binary input/output, but it gives good recognition results: this CNN chip with a simple 3 * 3 kernel can reliably classify 5 natural textures. Using more templates for decision-making, more textures can be separated and a classification error of less than 1% has been achieved.

AB - We have developed a new single-chip texture classifier smart-sensor system. Its main part is a programmable cellular neural network (CNN) VLSI chip with optical input and an execution time of a few microseconds. This chip contains a dynamic and locally interconnected 2D cell array. It executes a theoretically new method for texture classification, compared to the other convolution-based feature extraction methods, since here we have feedback convolution as well. Depending on the kernel parameters, this array can execute filtering, moving, linear and nonlinear effects at the same time. The parameters of the feedback and feed-forward convolutions are optimized through a genetic learning using the 22 * 20 CNN chip itself. This chip has a simplified architecture with binary input/output, but it gives good recognition results: this CNN chip with a simple 3 * 3 kernel can reliably classify 5 natural textures. Using more templates for decision-making, more textures can be separated and a classification error of less than 1% has been achieved.

KW - Cellular neural network

KW - Deconvolution

KW - Genetic algorithm

KW - Smart sensors

KW - Texture analysis

UR - http://www.scopus.com/inward/record.url?scp=0031274119&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031274119&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0031274119

VL - 18

SP - 1329

EP - 1334

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 11-13

ER -