Classification of Sonar Targets in Air—A Neural Network Approach

Classification of Sonar Targets in Air—A Neural Network Approach

Show other versions (1)

Ultrasonic sonar sensors are commonly used for contactless distance measurements in application areas such as automotive and mobile robotics. They can also be used to identify and classify sound reflecting objects. In the presented work, we classify simple sonar targets of different geometric shape...

Full description

Bibliographic Details
Main Authors: Patrick K. Kroh, Ralph Simon, Stefan J. Rupitsch
Format: Article
Language:English
Published: MDPI AG 2018-11-01
Series:Proceedings
Subjects:
sonar measurements
sonar detection
neural networks
feature extraction
Online Access:https://www.mdpi.com/2504-3900/2/13/929
id doaj-35164e22023644989812c89e78b42713
record_format Article
spelling doaj-35164e22023644989812c89e78b427132020-11-25T00:56:46ZengMDPI AGProceedings2504-39002018-11-0121392910.3390/proceedings2130929proceedings2130929Classification of Sonar Targets in Air—A Neural Network ApproachPatrick K. Kroh0Ralph Simon1Stefan J. Rupitsch2Lehrstuhl für Sensorik, Friedrich-Alexander-Universität Erlangen, 91052 Erlangen, GermanyDepartment of Ecological Science, Vrije Universiteit Amsterdam, 1081 HV Amsterdam, The NetherlandsLehrstuhl für Sensorik, Friedrich-Alexander-Universität Erlangen, 91052 Erlangen, GermanyUltrasonic sonar sensors are commonly used for contactless distance measurements in application areas such as automotive and mobile robotics. They can also be used to identify and classify sound reflecting objects. In the presented work, we classify simple sonar targets of different geometric shape and size. For this purpose, we built a test stand for echo measurements that facilitates defined arbitrary translation and rotation of the targets. Artificial neural networks (ANNs) with multiple hidden layers were used as classifiers and different features were evaluated. The focus was on two features that were derived from the echoes’ cross-correlation functions with their excitation chirp signals. We could distinguish different target geometries with our features and also evaluated the ANNs’ capabilities for size discrimination of targets with the same geometric shape.https://www.mdpi.com/2504-3900/2/13/929sonar measurementssonar detectionneural networksfeature extraction
collection DOAJ
language English
format Article
sources DOAJ
author Patrick K. Kroh
Ralph Simon
Stefan J. Rupitsch
spellingShingle Patrick K. Kroh
Ralph Simon
Stefan J. Rupitsch
Classification of Sonar Targets in Air—A Neural Network Approach
Proceedings
sonar measurements
sonar detection
neural networks
feature extraction
author_facet Patrick K. Kroh
Ralph Simon
Stefan J. Rupitsch
author_sort Patrick K. Kroh
title Classification of Sonar Targets in Air—A Neural Network Approach
title_short Classification of Sonar Targets in Air—A Neural Network Approach
title_full Classification of Sonar Targets in Air—A Neural Network Approach
title_fullStr Classification of Sonar Targets in Air—A Neural Network Approach
title_full_unstemmed Classification of Sonar Targets in Air—A Neural Network Approach
title_sort classification of sonar targets in air—a neural network approach
publisher MDPI AG
series Proceedings
issn 2504-3900
publishDate 2018-11-01
description Ultrasonic sonar sensors are commonly used for contactless distance measurements in application areas such as automotive and mobile robotics. They can also be used to identify and classify sound reflecting objects. In the presented work, we classify simple sonar targets of different geometric shape and size. For this purpose, we built a test stand for echo measurements that facilitates defined arbitrary translation and rotation of the targets. Artificial neural networks (ANNs) with multiple hidden layers were used as classifiers and different features were evaluated. The focus was on two features that were derived from the echoes’ cross-correlation functions with their excitation chirp signals. We could distinguish different target geometries with our features and also evaluated the ANNs’ capabilities for size discrimination of targets with the same geometric shape.
topic sonar measurements
sonar detection
neural networks
feature extraction
url https://www.mdpi.com/2504-3900/2/13/929
work_keys_str_mv AT patrickkkroh classificationofsonartargetsinairaneuralnetworkapproach
AT ralphsimon classificationofsonartargetsinairaneuralnetworkapproach
AT stefanjrupitsch classificationofsonartargetsinairaneuralnetworkapproach
_version_ 1725225581754187776

Similar Items