Feature extraction technique using discrete wavelet transform for image classification

Research output: Chapter in Book/Report/Conference proceedingConference contribution

19 Citations (Scopus)

Abstract

The purpose of feature extraction technique in image processing is to represent the image in its compact and unique form of single values or matrix vector. Low level feature extraction involves automatic extraction of features from an image without doing any processing method. In this paper, we consider the use of high level feature extraction technique to investigate the characteristic of narrow and broad weed by implementing the 2 Dimensional Discrete Wavelet Transform (2D-DWT) as the processing method. Most transformation techniques produce coefficient values with the same size as the original image. Further processing of the coefficient values must be applied to extract the image feature vectors. In this paper, we propose an algorithm to implement feature extraction technique using the 2D-DWT and the extracted coefficients are used to represent the image for classification of narrow and broad weed. Results obtained suggest that the extracted 2D-DWT coefficients can uniquely represents the two different weed type.

Original languageEnglish
Title of host publication2007 5th Student Conference on Research and Development, SCORED
DOIs
Publication statusPublished - 2007
Event2007 5th Student Conference on Research and Development, SCORED - Selangor
Duration: 11 Dec 200712 Dec 2007

Other

Other2007 5th Student Conference on Research and Development, SCORED
CitySelangor
Period11/12/0712/12/07

Fingerprint

Values
Wavelet transform
Feature extraction
Coefficients
Weeds
Image processing

Keywords

  • Coefficient
  • Discrete wavelet transform
  • Feature extraction
  • Weed

ASJC Scopus subject areas

  • Education
  • Management Science and Operations Research

Cite this

Feature extraction technique using discrete wavelet transform for image classification. / Ghazali, Kamarul Hawari; Mansor, Mohd Fais; Mustafa, Mohd. Marzuki; Hussain, Aini.

2007 5th Student Conference on Research and Development, SCORED. 2007. 4451366.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ghazali, KH, Mansor, MF, Mustafa, MM & Hussain, A 2007, Feature extraction technique using discrete wavelet transform for image classification. in 2007 5th Student Conference on Research and Development, SCORED., 4451366, 2007 5th Student Conference on Research and Development, SCORED, Selangor, 11/12/07. https://doi.org/10.1109/SCORED.2007.4451366
Ghazali, Kamarul Hawari ; Mansor, Mohd Fais ; Mustafa, Mohd. Marzuki ; Hussain, Aini. / Feature extraction technique using discrete wavelet transform for image classification. 2007 5th Student Conference on Research and Development, SCORED. 2007.
@inproceedings{d11658c7786c41e1a8d2a9bcd8e64d3b,
title = "Feature extraction technique using discrete wavelet transform for image classification",
abstract = "The purpose of feature extraction technique in image processing is to represent the image in its compact and unique form of single values or matrix vector. Low level feature extraction involves automatic extraction of features from an image without doing any processing method. In this paper, we consider the use of high level feature extraction technique to investigate the characteristic of narrow and broad weed by implementing the 2 Dimensional Discrete Wavelet Transform (2D-DWT) as the processing method. Most transformation techniques produce coefficient values with the same size as the original image. Further processing of the coefficient values must be applied to extract the image feature vectors. In this paper, we propose an algorithm to implement feature extraction technique using the 2D-DWT and the extracted coefficients are used to represent the image for classification of narrow and broad weed. Results obtained suggest that the extracted 2D-DWT coefficients can uniquely represents the two different weed type.",
keywords = "Coefficient, Discrete wavelet transform, Feature extraction, Weed",
author = "Ghazali, {Kamarul Hawari} and Mansor, {Mohd Fais} and Mustafa, {Mohd. Marzuki} and Aini Hussain",
year = "2007",
doi = "10.1109/SCORED.2007.4451366",
language = "English",
isbn = "1424414709",
booktitle = "2007 5th Student Conference on Research and Development, SCORED",

}

TY - GEN

T1 - Feature extraction technique using discrete wavelet transform for image classification

AU - Ghazali, Kamarul Hawari

AU - Mansor, Mohd Fais

AU - Mustafa, Mohd. Marzuki

AU - Hussain, Aini

PY - 2007

Y1 - 2007

N2 - The purpose of feature extraction technique in image processing is to represent the image in its compact and unique form of single values or matrix vector. Low level feature extraction involves automatic extraction of features from an image without doing any processing method. In this paper, we consider the use of high level feature extraction technique to investigate the characteristic of narrow and broad weed by implementing the 2 Dimensional Discrete Wavelet Transform (2D-DWT) as the processing method. Most transformation techniques produce coefficient values with the same size as the original image. Further processing of the coefficient values must be applied to extract the image feature vectors. In this paper, we propose an algorithm to implement feature extraction technique using the 2D-DWT and the extracted coefficients are used to represent the image for classification of narrow and broad weed. Results obtained suggest that the extracted 2D-DWT coefficients can uniquely represents the two different weed type.

AB - The purpose of feature extraction technique in image processing is to represent the image in its compact and unique form of single values or matrix vector. Low level feature extraction involves automatic extraction of features from an image without doing any processing method. In this paper, we consider the use of high level feature extraction technique to investigate the characteristic of narrow and broad weed by implementing the 2 Dimensional Discrete Wavelet Transform (2D-DWT) as the processing method. Most transformation techniques produce coefficient values with the same size as the original image. Further processing of the coefficient values must be applied to extract the image feature vectors. In this paper, we propose an algorithm to implement feature extraction technique using the 2D-DWT and the extracted coefficients are used to represent the image for classification of narrow and broad weed. Results obtained suggest that the extracted 2D-DWT coefficients can uniquely represents the two different weed type.

KW - Coefficient

KW - Discrete wavelet transform

KW - Feature extraction

KW - Weed

UR - http://www.scopus.com/inward/record.url?scp=50449089280&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50449089280&partnerID=8YFLogxK

U2 - 10.1109/SCORED.2007.4451366

DO - 10.1109/SCORED.2007.4451366

M3 - Conference contribution

SN - 1424414709

SN - 9781424414703

BT - 2007 5th Student Conference on Research and Development, SCORED

ER -