Fish detection based on histogram of image derivative for hauling operations

C. H. Kuo, H. H. Lin, Chung-Hung Lin, Z. H. Lin

Research output: Contribution to journalArticle

Abstract

This study attempts to explore the methods to identify fishes in still images taken from the electronic observer system installed on longline fishing vessels. Sobel filter was used to capture image features and transform it into histogram. Different recognition mechanisms were tested to find out the best method. Captured image features include: gradient magnitude, angle, and curvature. Correlation coefficient was chosen as baseline of recognition mechanisms. K-means clustering method was used; however, the results showed the effectiveness of recognition is not as expected. Thus, "fish global feature" and "fish local feature" were proposed. First, a single tuna template was created and tested. After the validity was established, by combining more templates and utilizing Bagging algorithm, the recognition accuracy can be improved. In the experiment, 96.0% fish recognition rate was obtained. The results show that adequate number of templates were created with corresponding with site conditions, the recognition accuracy can be increased.

Original languageEnglish
Pages (from-to)135-144
Number of pages10
JournalJournal of Taiwan Society of Naval Architects and Marine Engineers
Volume35
Issue number3
Publication statusPublished - 2016 Aug 1

Fingerprint

Fish
Derivatives
Fishing vessels
Experiments

All Science Journal Classification (ASJC) codes

  • Ocean Engineering
  • Mechanical Engineering

Cite this

@article{36e26fb66ab04c93818e1a1d5f79c43e,
title = "Fish detection based on histogram of image derivative for hauling operations",
abstract = "This study attempts to explore the methods to identify fishes in still images taken from the electronic observer system installed on longline fishing vessels. Sobel filter was used to capture image features and transform it into histogram. Different recognition mechanisms were tested to find out the best method. Captured image features include: gradient magnitude, angle, and curvature. Correlation coefficient was chosen as baseline of recognition mechanisms. K-means clustering method was used; however, the results showed the effectiveness of recognition is not as expected. Thus, {"}fish global feature{"} and {"}fish local feature{"} were proposed. First, a single tuna template was created and tested. After the validity was established, by combining more templates and utilizing Bagging algorithm, the recognition accuracy can be improved. In the experiment, 96.0{\%} fish recognition rate was obtained. The results show that adequate number of templates were created with corresponding with site conditions, the recognition accuracy can be increased.",
author = "Kuo, {C. H.} and Lin, {H. H.} and Chung-Hung Lin and Lin, {Z. H.}",
year = "2016",
month = "8",
day = "1",
language = "English",
volume = "35",
pages = "135--144",
journal = "Journal of Taiwan Society of Naval Architects and Marine Engineers",
issn = "1023-4535",
publisher = "Taiwan Society of Naval Architects and Marine Engineers",
number = "3",

}

Fish detection based on histogram of image derivative for hauling operations. / Kuo, C. H.; Lin, H. H.; Lin, Chung-Hung; Lin, Z. H.

In: Journal of Taiwan Society of Naval Architects and Marine Engineers, Vol. 35, No. 3, 01.08.2016, p. 135-144.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fish detection based on histogram of image derivative for hauling operations

AU - Kuo, C. H.

AU - Lin, H. H.

AU - Lin, Chung-Hung

AU - Lin, Z. H.

PY - 2016/8/1

Y1 - 2016/8/1

N2 - This study attempts to explore the methods to identify fishes in still images taken from the electronic observer system installed on longline fishing vessels. Sobel filter was used to capture image features and transform it into histogram. Different recognition mechanisms were tested to find out the best method. Captured image features include: gradient magnitude, angle, and curvature. Correlation coefficient was chosen as baseline of recognition mechanisms. K-means clustering method was used; however, the results showed the effectiveness of recognition is not as expected. Thus, "fish global feature" and "fish local feature" were proposed. First, a single tuna template was created and tested. After the validity was established, by combining more templates and utilizing Bagging algorithm, the recognition accuracy can be improved. In the experiment, 96.0% fish recognition rate was obtained. The results show that adequate number of templates were created with corresponding with site conditions, the recognition accuracy can be increased.

AB - This study attempts to explore the methods to identify fishes in still images taken from the electronic observer system installed on longline fishing vessels. Sobel filter was used to capture image features and transform it into histogram. Different recognition mechanisms were tested to find out the best method. Captured image features include: gradient magnitude, angle, and curvature. Correlation coefficient was chosen as baseline of recognition mechanisms. K-means clustering method was used; however, the results showed the effectiveness of recognition is not as expected. Thus, "fish global feature" and "fish local feature" were proposed. First, a single tuna template was created and tested. After the validity was established, by combining more templates and utilizing Bagging algorithm, the recognition accuracy can be improved. In the experiment, 96.0% fish recognition rate was obtained. The results show that adequate number of templates were created with corresponding with site conditions, the recognition accuracy can be increased.

UR - http://www.scopus.com/inward/record.url?scp=85014075604&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85014075604&partnerID=8YFLogxK

M3 - Article

VL - 35

SP - 135

EP - 144

JO - Journal of Taiwan Society of Naval Architects and Marine Engineers

JF - Journal of Taiwan Society of Naval Architects and Marine Engineers

SN - 1023-4535

IS - 3

ER -