Classification and retrieval on macroinvertebrate image databases

Serkan Kiranyaz, Turker Ince, Jenni Pulkkinen, Moncef Gabbouj, Johanna Ärje, Salme Kärkkäinen, Ville Tirronen, Martti Juhola, Tuomas Turpeinen, Kristian Meissner

Research output: Contribution to journalArticleScientificpeer-review

21 Citations (Scopus)

Abstract

Aquatic ecosystems are continuously threatened by a growing number of human induced changes. Macroinvertebrate biomonitoring is particularly efficient in pinpointing the cause-effect structure between slow and subtle changes and their detrimental consequences in aquatic ecosystems. The greatest obstacle to implementing efficient biomonitoring is currently the cost-intensive human expert taxonomic identification of samples. While there is evidence that automated recognition techniques can match human taxa identification accuracy at greatly reduced costs, so far the development of automated identification techniques for aquatic organisms has been minimal. In this paper, we focus on advancing classification and data retrieval that are instrumental when processing large macroinvertebrate image datasets. To accomplish this for routine biomonitoring, in this paper we shall investigate the feasibility of automated river macroinvertebrate classification and retrieval with high precision. Besides the state-of-the-art classifiers such as Support Vector Machines (SVMs) and Bayesian Classifiers (BCs), the focus is particularly drawn on feed-forward artificial neural networks (ANNs), namely multilayer perceptrons (MLPs) and radial basis function networks (RBFNs). Since both ANN types have been proclaimed superior by different investigations even for the same benchmark problems, we shall first show that the main reason for this ambiguity lies in the static and rather poor comparison methodologies applied in most earlier works. Especially the most common drawback occurs due to the limited evaluation of the ANN performances over just one or few network architecture(s). Therefore, in this study, an extensive evaluation of each classifier performance over an ANN architecture space is performed. The best classifier among all, which is trained over a dataset of river macroinvertebrate specimens, is then used in the MUVIS framework for the efficient search and retrieval of particular macroinvertebrate peculiars. Classification and retrieval results present high accuracy and can match an experts ability for taxonomic identification. © 2011 Elsevier Ltd.
Original languageEnglish
Pages (from-to)463-472
JournalComputers in Biology and Medicine
Volume41
Issue number7
DOIs
Publication statusPublished - 2011
MoE publication typeNot Eligible

Fingerprint

Environmental Monitoring
Classifiers
Databases
Neural networks
Rivers
Aquatic ecosystems
Ecosystem
Network architecture
Costs and Cost Analysis
Benchmarking
Aquatic Organisms
Forensic Anthropology
Neural Networks (Computer)
Information Storage and Retrieval
Aquatic organisms
Radial basis function networks
Multilayer neural networks
Network performance
Support vector machines
Costs

Keywords

  • Bayesian Networks
  • Benthic macroinvertebrate
  • Biomonitoring
  • Classification
  • Multilayer perceptrons
  • Radial basis function networks
  • Support Vector Machines

Cite this

Kiranyaz, S., Ince, T., Pulkkinen, J., Gabbouj, M., Ärje, J., Kärkkäinen, S., ... Meissner, K. (2011). Classification and retrieval on macroinvertebrate image databases. Computers in Biology and Medicine, 41(7), 463-472. https://doi.org/10.1016/j.compbiomed.2011.04.008
Kiranyaz, Serkan ; Ince, Turker ; Pulkkinen, Jenni ; Gabbouj, Moncef ; Ärje, Johanna ; Kärkkäinen, Salme ; Tirronen, Ville ; Juhola, Martti ; Turpeinen, Tuomas ; Meissner, Kristian. / Classification and retrieval on macroinvertebrate image databases. In: Computers in Biology and Medicine. 2011 ; Vol. 41, No. 7. pp. 463-472.
@article{a2a0df66ad9f4a8eb43179871d62e5a4,
title = "Classification and retrieval on macroinvertebrate image databases",
abstract = "Aquatic ecosystems are continuously threatened by a growing number of human induced changes. Macroinvertebrate biomonitoring is particularly efficient in pinpointing the cause-effect structure between slow and subtle changes and their detrimental consequences in aquatic ecosystems. The greatest obstacle to implementing efficient biomonitoring is currently the cost-intensive human expert taxonomic identification of samples. While there is evidence that automated recognition techniques can match human taxa identification accuracy at greatly reduced costs, so far the development of automated identification techniques for aquatic organisms has been minimal. In this paper, we focus on advancing classification and data retrieval that are instrumental when processing large macroinvertebrate image datasets. To accomplish this for routine biomonitoring, in this paper we shall investigate the feasibility of automated river macroinvertebrate classification and retrieval with high precision. Besides the state-of-the-art classifiers such as Support Vector Machines (SVMs) and Bayesian Classifiers (BCs), the focus is particularly drawn on feed-forward artificial neural networks (ANNs), namely multilayer perceptrons (MLPs) and radial basis function networks (RBFNs). Since both ANN types have been proclaimed superior by different investigations even for the same benchmark problems, we shall first show that the main reason for this ambiguity lies in the static and rather poor comparison methodologies applied in most earlier works. Especially the most common drawback occurs due to the limited evaluation of the ANN performances over just one or few network architecture(s). Therefore, in this study, an extensive evaluation of each classifier performance over an ANN architecture space is performed. The best classifier among all, which is trained over a dataset of river macroinvertebrate specimens, is then used in the MUVIS framework for the efficient search and retrieval of particular macroinvertebrate peculiars. Classification and retrieval results present high accuracy and can match an experts ability for taxonomic identification. {\circledC} 2011 Elsevier Ltd.",
keywords = "Bayesian Networks, Benthic macroinvertebrate, Biomonitoring, Classification, Multilayer perceptrons, Radial basis function networks, Support Vector Machines",
author = "Serkan Kiranyaz and Turker Ince and Jenni Pulkkinen and Moncef Gabbouj and Johanna {\"A}rje and Salme K{\"a}rkk{\"a}inen and Ville Tirronen and Martti Juhola and Tuomas Turpeinen and Kristian Meissner",
year = "2011",
doi = "10.1016/j.compbiomed.2011.04.008",
language = "English",
volume = "41",
pages = "463--472",
journal = "Computers in Biology and Medicine",
issn = "0010-4825",
publisher = "Elsevier",
number = "7",

}

Kiranyaz, S, Ince, T, Pulkkinen, J, Gabbouj, M, Ärje, J, Kärkkäinen, S, Tirronen, V, Juhola, M, Turpeinen, T & Meissner, K 2011, 'Classification and retrieval on macroinvertebrate image databases', Computers in Biology and Medicine, vol. 41, no. 7, pp. 463-472. https://doi.org/10.1016/j.compbiomed.2011.04.008

Classification and retrieval on macroinvertebrate image databases. / Kiranyaz, Serkan; Ince, Turker; Pulkkinen, Jenni; Gabbouj, Moncef; Ärje, Johanna; Kärkkäinen, Salme; Tirronen, Ville; Juhola, Martti; Turpeinen, Tuomas; Meissner, Kristian.

In: Computers in Biology and Medicine, Vol. 41, No. 7, 2011, p. 463-472.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Classification and retrieval on macroinvertebrate image databases

AU - Kiranyaz, Serkan

AU - Ince, Turker

AU - Pulkkinen, Jenni

AU - Gabbouj, Moncef

AU - Ärje, Johanna

AU - Kärkkäinen, Salme

AU - Tirronen, Ville

AU - Juhola, Martti

AU - Turpeinen, Tuomas

AU - Meissner, Kristian

PY - 2011

Y1 - 2011

N2 - Aquatic ecosystems are continuously threatened by a growing number of human induced changes. Macroinvertebrate biomonitoring is particularly efficient in pinpointing the cause-effect structure between slow and subtle changes and their detrimental consequences in aquatic ecosystems. The greatest obstacle to implementing efficient biomonitoring is currently the cost-intensive human expert taxonomic identification of samples. While there is evidence that automated recognition techniques can match human taxa identification accuracy at greatly reduced costs, so far the development of automated identification techniques for aquatic organisms has been minimal. In this paper, we focus on advancing classification and data retrieval that are instrumental when processing large macroinvertebrate image datasets. To accomplish this for routine biomonitoring, in this paper we shall investigate the feasibility of automated river macroinvertebrate classification and retrieval with high precision. Besides the state-of-the-art classifiers such as Support Vector Machines (SVMs) and Bayesian Classifiers (BCs), the focus is particularly drawn on feed-forward artificial neural networks (ANNs), namely multilayer perceptrons (MLPs) and radial basis function networks (RBFNs). Since both ANN types have been proclaimed superior by different investigations even for the same benchmark problems, we shall first show that the main reason for this ambiguity lies in the static and rather poor comparison methodologies applied in most earlier works. Especially the most common drawback occurs due to the limited evaluation of the ANN performances over just one or few network architecture(s). Therefore, in this study, an extensive evaluation of each classifier performance over an ANN architecture space is performed. The best classifier among all, which is trained over a dataset of river macroinvertebrate specimens, is then used in the MUVIS framework for the efficient search and retrieval of particular macroinvertebrate peculiars. Classification and retrieval results present high accuracy and can match an experts ability for taxonomic identification. © 2011 Elsevier Ltd.

AB - Aquatic ecosystems are continuously threatened by a growing number of human induced changes. Macroinvertebrate biomonitoring is particularly efficient in pinpointing the cause-effect structure between slow and subtle changes and their detrimental consequences in aquatic ecosystems. The greatest obstacle to implementing efficient biomonitoring is currently the cost-intensive human expert taxonomic identification of samples. While there is evidence that automated recognition techniques can match human taxa identification accuracy at greatly reduced costs, so far the development of automated identification techniques for aquatic organisms has been minimal. In this paper, we focus on advancing classification and data retrieval that are instrumental when processing large macroinvertebrate image datasets. To accomplish this for routine biomonitoring, in this paper we shall investigate the feasibility of automated river macroinvertebrate classification and retrieval with high precision. Besides the state-of-the-art classifiers such as Support Vector Machines (SVMs) and Bayesian Classifiers (BCs), the focus is particularly drawn on feed-forward artificial neural networks (ANNs), namely multilayer perceptrons (MLPs) and radial basis function networks (RBFNs). Since both ANN types have been proclaimed superior by different investigations even for the same benchmark problems, we shall first show that the main reason for this ambiguity lies in the static and rather poor comparison methodologies applied in most earlier works. Especially the most common drawback occurs due to the limited evaluation of the ANN performances over just one or few network architecture(s). Therefore, in this study, an extensive evaluation of each classifier performance over an ANN architecture space is performed. The best classifier among all, which is trained over a dataset of river macroinvertebrate specimens, is then used in the MUVIS framework for the efficient search and retrieval of particular macroinvertebrate peculiars. Classification and retrieval results present high accuracy and can match an experts ability for taxonomic identification. © 2011 Elsevier Ltd.

KW - Bayesian Networks

KW - Benthic macroinvertebrate

KW - Biomonitoring

KW - Classification

KW - Multilayer perceptrons

KW - Radial basis function networks

KW - Support Vector Machines

U2 - 10.1016/j.compbiomed.2011.04.008

DO - 10.1016/j.compbiomed.2011.04.008

M3 - Article

VL - 41

SP - 463

EP - 472

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

SN - 0010-4825

IS - 7

ER -

Kiranyaz S, Ince T, Pulkkinen J, Gabbouj M, Ärje J, Kärkkäinen S et al. Classification and retrieval on macroinvertebrate image databases. Computers in Biology and Medicine. 2011;41(7):463-472. https://doi.org/10.1016/j.compbiomed.2011.04.008