Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks

Helmut Tödtmann; Matthias Vahl; Uwe von Lukas; Torsten Ullrich

doi:10.5220/0008962803700377

Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks

Helmut Tödtmann, Matthias Vahl, Uwe von Lukas, Torsten Ullrich

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Abstract

Monitoring the environment for early recognition of changes is necessary for assessing the success of renaturation measures on a facts basis. It is also used in fisheries and livestock production for monitoring and for quality assurance. The goal of the presented system is to count sea trouts annually over the course of several months. Sea trouts are detected with underwater camera systems triggered by motion sensors. Such a scenario generates many videos that have to be evaluated manually. This article describes the techniques used to automate the image evaluation process. An effective method has been developed to classify videos and determine the times of occurrence of sea trouts, while significantly reducing the annotation effort. A convolutional neural network has been trained via supervised learning. The underlying images are frame compositions automatically extracted from videos on which sea trouts are to be detected. The accuracy of the resulting detection system reaches values of up to 97.7 %.

Original language	English
Title of host publication	Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Editors	Giovanni Maria Farinella, Petia Radeva, Jose Braz
Publisher	SciTePress
Pages	370-377
Number of pages	8
Volume	5, VISAPP
ISBN (Electronic)	978-989-758-402-2
DOIs	https://doi.org/10.5220/0008962803700377
Publication status	Published - 2020
Event	16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications: VISIGRAPP 2021 - Virtuell, Austria Duration: 8 Feb 2021 → 10 Feb 2021

Conference

Conference	16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Country/Territory	Austria
City	Virtuell
Period	8/02/21 → 10/02/21

Keywords

Convolutional Neural Network
Deep Learning
Detection
Environmental Monitoring
Implicit Segmentation

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Computer Science Applications
Computer Graphics and Computer-Aided Design

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.5220/0008962803700377Licence: CC BY-NC-ND 4.0

Cite this

Tödtmann, H., Vahl, M., von Lukas, U., & Ullrich, T. (2020). Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks. In G. M. Farinella, P. Radeva, & J. Braz (Eds.), Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Vol. 5, VISAPP, pp. 370-377). SciTePress. https://doi.org/10.5220/0008962803700377

Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks. / Tödtmann, Helmut; Vahl, Matthias; von Lukas, Uwe et al.
Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . ed. / Giovanni Maria Farinella; Petia Radeva; Jose Braz. Vol. 5, VISAPP SciTePress, 2020. p. 370-377.

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Tödtmann, H, Vahl, M, von Lukas, U & Ullrich, T 2020, Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks. in GM Farinella, P Radeva & J Braz (eds), Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . vol. 5, VISAPP, SciTePress, pp. 370-377, 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Virtuell, Austria, 8/02/21. https://doi.org/10.5220/0008962803700377

Tödtmann H, Vahl M, von Lukas U, Ullrich T. Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks. In Farinella GM, Radeva P, Braz J, editors, Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . Vol. 5, VISAPP. SciTePress. 2020. p. 370-377 doi: 10.5220/0008962803700377

Tödtmann, Helmut ; Vahl, Matthias ; von Lukas, Uwe et al. / Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks. Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . editor / Giovanni Maria Farinella ; Petia Radeva ; Jose Braz. Vol. 5, VISAPP SciTePress, 2020. pp. 370-377

@inproceedings{85bb393c22e34064900be25163260471,

title = "Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks",

abstract = "Monitoring the environment for early recognition of changes is necessary for assessing the success of renaturation measures on a facts basis. It is also used in fisheries and livestock production for monitoring and for quality assurance. The goal of the presented system is to count sea trouts annually over the course of several months. Sea trouts are detected with underwater camera systems triggered by motion sensors. Such a scenario generates many videos that have to be evaluated manually. This article describes the techniques used to automate the image evaluation process. An effective method has been developed to classify videos and determine the times of occurrence of sea trouts, while significantly reducing the annotation effort. A convolutional neural network has been trained via supervised learning. The underlying images are frame compositions automatically extracted from videos on which sea trouts are to be detected. The accuracy of the resulting detection system reaches values of up to 97.7 %.",

keywords = "Convolutional Neural Network, Deep Learning, Detection, Environmental Monitoring, Implicit Segmentation",

author = "Helmut T{\"o}dtmann and Matthias Vahl and {von Lukas}, Uwe and Torsten Ullrich",

year = "2020",

doi = "10.5220/0008962803700377",

language = "English",

volume = "5, VISAPP",

pages = "370--377",

editor = "Farinella, {Giovanni Maria} and Petia Radeva and Jose Braz",

booktitle = "Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications",

publisher = "SciTePress",

address = "Portugal",

note = "16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : VISIGRAPP 2021 ; Conference date: 08-02-2021 Through 10-02-2021",

}

TY - GEN

T1 - Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks

AU - Tödtmann, Helmut

AU - Vahl, Matthias

AU - von Lukas, Uwe

AU - Ullrich, Torsten

PY - 2020

Y1 - 2020

N2 - Monitoring the environment for early recognition of changes is necessary for assessing the success of renaturation measures on a facts basis. It is also used in fisheries and livestock production for monitoring and for quality assurance. The goal of the presented system is to count sea trouts annually over the course of several months. Sea trouts are detected with underwater camera systems triggered by motion sensors. Such a scenario generates many videos that have to be evaluated manually. This article describes the techniques used to automate the image evaluation process. An effective method has been developed to classify videos and determine the times of occurrence of sea trouts, while significantly reducing the annotation effort. A convolutional neural network has been trained via supervised learning. The underlying images are frame compositions automatically extracted from videos on which sea trouts are to be detected. The accuracy of the resulting detection system reaches values of up to 97.7 %.

AB - Monitoring the environment for early recognition of changes is necessary for assessing the success of renaturation measures on a facts basis. It is also used in fisheries and livestock production for monitoring and for quality assurance. The goal of the presented system is to count sea trouts annually over the course of several months. Sea trouts are detected with underwater camera systems triggered by motion sensors. Such a scenario generates many videos that have to be evaluated manually. This article describes the techniques used to automate the image evaluation process. An effective method has been developed to classify videos and determine the times of occurrence of sea trouts, while significantly reducing the annotation effort. A convolutional neural network has been trained via supervised learning. The underlying images are frame compositions automatically extracted from videos on which sea trouts are to be detected. The accuracy of the resulting detection system reaches values of up to 97.7 %.

KW - Convolutional Neural Network

KW - Deep Learning

KW - Detection

KW - Environmental Monitoring

KW - Implicit Segmentation

UR - http://www.scopus.com/inward/record.url?scp=85083493724&partnerID=8YFLogxK

U2 - 10.5220/0008962803700377

DO - 10.5220/0008962803700377

M3 - Conference paper

VL - 5, VISAPP

SP - 370

EP - 377

BT - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

A2 - Farinella, Giovanni Maria

A2 - Radeva, Petia

A2 - Braz, Jose

PB - SciTePress

T2 - 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

Y2 - 8 February 2021 through 10 February 2021

ER -

Time-Unfolding Object Existence Detection in Low-Quality Underwater Videos using Convolutional Neural Networks

Abstract

Conference

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this