<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>1820</id>
    <completedYear>2022</completedYear>
    <publishedYear>2022</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>4</pageNumber>
    <edition/>
    <issue/>
    <volume>2022</volume>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">An Acoustical Machine Learning Approach to Determine Abrasive Belt Wear of Wide Belt Sanders</title>
    <abstract language="eng">This paper describes a machine learning approach to determine the abrasive belt wear of wide belt sanders used in industrial processes based on acoustic data, regardless of the sanding process-related parameters, Feed speed, Grit Size, and Type of material. Our approach utilizes Decision Tree, Random Forest, k-nearest Neighbors, and Neural network Classifiers to detect the belt wear from Spectrograms, Mel Spectrograms, MFCC, IMFCC, and LFCC, yielding an accuracy of up to 86.1% on five levels of belt wear. A 96% accuracy could be achieved with different Decision Tree Classifiers specialized in different sanding parameter configurations. The classifiers could also determine with an accuracy of 97% if the machine is currently sanding or is idle and with an accuracy of 98.4% and 98.8% detect the sanding parameters Feed speed and Grit Size. We can show that low-dimensional mappings of high-dimensional features can be used to visualize belt wear and sanding parameters meaningfully.</abstract>
    <parentTitle language="eng">2022 IEEE Sensors</parentTitle>
    <identifier type="isbn">978-1-6654-8464-0</identifier>
    <identifier type="doi">10.1109/SENSORS52175.2022.9967324</identifier>
    <enrichment key="opus_doi_flag">true</enrichment>
    <enrichment key="opus_import_data">{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,6,25]],"date-time":"2024-06-25T06:00:00Z","timestamp":1719295200117},"reference-count":16,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,10,30]],"date-time":"2022-10-30T00:00:00Z","timestamp":1667088000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,10,30]],"date-time":"2022-10-30T00:00:00Z","timestamp":1667088000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100004895","name":"European Social Funds (ESF)","doi-asserted-by":"publisher","award":["R.6-V0332.2.43\/1\/5"]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,10,30]]},"DOI":"10.1109\/sensors52175.2022.9967324","type":"proceedings-article","created":{"date-parts":[[2022,12,8]],"date-time":"2022-12-08T18:42:41Z","timestamp":1670524961000},"source":"Crossref","is-referenced-by-count":1,"title":["An Acoustical Machine Learning Approach to Determine Abrasive Belt Wear of Wide Belt Sanders"],"prefix":"10.1109","author":[{"given":"Maximilian","family":"Bundscherer","sequence":"first","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,N&amp;#x00FC;rnberg,Germany"}]},{"given":"Thomas H.","family":"Schmitt","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,N&amp;#x00FC;rnberg,Germany"}]},{"given":"Sebastian","family":"Bayerl","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,N&amp;#x00FC;rnberg,Germany"}]},{"given":"Thomas","family":"Auerbach","sequence":"additional","affiliation":[{"name":"Production Technology, Hans Weber Maschinenfabrik GmbH,Kronach,Germany"}]},{"given":"Tobias","family":"Bocklet","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,N&amp;#x00FC;rnberg,Germany"}]}],"member":"263","reference":[{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmapro.2017.09.030"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1999.758101"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1007\/BF02943243"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/LA-CCI.2018.8625230"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICSIGSYS.2019.8811070"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2019.03.037"},{"key":"ref16","first-page":"173","year":"2007","journal-title":"Measuring the Performance of a Classifier"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.triboint.2018.01.019"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.3390\/electronics10121429"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/s00170-017-0498-4"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1115\/1.4002022"},{"key":"ref8","first-page":"80","article-title":"Recording electric guitar&amp;#x2014;the science and the myth","volume":"58","author":"case","year":"2010","journal-title":"J Audio Eng Soc"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/EDUCON.2015.7096011"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1016\/S0924-0136(01)00621-5"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmapro.2019.12.034"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.3390\/s101110401"}],"event":{"name":"2022 IEEE Sensors","location":"Dallas, TX, USA","start":{"date-parts":[[2022,10,30]]},"end":{"date-parts":[[2022,11,2]]}},"container-title":["2022 IEEE Sensors"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9966996\/9966997\/09967324.pdf?arnumber=9967324","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,26]],"date-time":"2022-12-26T19:47:22Z","timestamp":1672084042000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9967324\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,30]]},"references-count":16,"URL":"http:\/\/dx.doi.org\/10.1109\/sensors52175.2022.9967324","relation":{},"subject":[],"published":{"date-parts":[[2022,10,30]]}}}</enrichment>
    <enrichment key="local_crossrefDocumentType">proceedings-article</enrichment>
    <enrichment key="local_crossrefLicence">https://doi.org/10.15223/policy-029</enrichment>
    <enrichment key="local_import_origin">crossref</enrichment>
    <enrichment key="local_doiImportPopulated">PersonAuthorFirstName_1,PersonAuthorLastName_1,PersonAuthorFirstName_2,PersonAuthorLastName_2,PersonAuthorFirstName_3,PersonAuthorLastName_3,PersonAuthorFirstName_4,PersonAuthorLastName_4,PersonAuthorFirstName_5,PersonAuthorLastName_5,Enrichmentconference_title,Enrichmentconference_place,PublisherName,TitleMain_1,TitleParent_1,CompletedYear,Enrichmentlocal_crossrefLicence</enrichment>
    <enrichment key="conference_title">2022 IEEE Sensors</enrichment>
    <enrichment key="conference_place">Dallas, TX, USA</enrichment>
    <enrichment key="opus.source">doi-import</enrichment>
    <enrichment key="Reviewstatus">Begutachtet/Reviewed</enrichment>
    <enrichment key="ConferenceStatement">2022 IEEE Sensors, 30 October 2022 - 02 November 2022, Dallas, TX, USA</enrichment>
    <author>Maximilian Bundscherer</author>
    <author>Thomas H. Schmitt</author>
    <author>Sebastian P. Bayerl</author>
    <author>Thomas Auerbach</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Acoustic sensors</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Abrasive belt wear</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Tool wear</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Machine learning</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Industrial process</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Wide belt sanding machines</value>
    </subject>
    <collection role="institutes" number="">Fakultät Informatik</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
  </doc>
  <doc>
    <id>1823</id>
    <completedYear>2022</completedYear>
    <publishedYear>2022</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>4</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Machine learning based optimization of a ceramic bushing manufacturing process</title>
    <abstract language="eng">Machine learning (ML) has shown great promise in a variety of domains in recent years. ML models are known to require large amounts of labeled training data, keeping small to medium-sized business from utilizing them. This paper presents ML based approach to optimize a ceramic bushing manufac-turing process, by predicting the employed press-fit process as a function of press punch position. Accurate predictions would ensure optimal process configuration, guaranteeing quality and reducing waste. Models are trained in a supervised manner to predict the press-fit process and the ceramic defect probabilities as functions of press punch position. We were able to predict the press-fit process with a mean correlation of 0.996 and assess whether the process would damage the ceramic with a mean precision of 96.7%. Our results exemplify how ML can be used to predict and optimize highly specialised processes even with small datasets.</abstract>
    <parentTitle language="eng">2022 IEEE Sensors</parentTitle>
    <identifier type="isbn">978-1-6654-8464-0</identifier>
    <identifier type="doi">10.1109/sensors52175.2022.9967124</identifier>
    <enrichment key="opus_doi_flag">true</enrichment>
    <enrichment key="opus_import_data">{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2022,12,27]],"date-time":"2022-12-27T06:07:43Z","timestamp":1672121263787},"reference-count":14,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,10,30]],"date-time":"2022-10-30T00:00:00Z","timestamp":1667088000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,10,30]],"date-time":"2022-10-30T00:00:00Z","timestamp":1667088000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,10,30]]},"DOI":"10.1109\/sensors52175.2022.9967124","type":"proceedings-article","created":{"date-parts":[[2022,12,8]],"date-time":"2022-12-08T18:42:41Z","timestamp":1670524961000},"source":"Crossref","is-referenced-by-count":0,"title":["Machine learning based optimization of a ceramic bushing manufacturing process"],"prefix":"10.1109","author":[{"given":"Thomas H.","family":"Schmitt","sequence":"first","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,Nurnberg,Germany"}]},{"given":"Maximilian","family":"Bundscherer","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,Nurnberg,Germany"}]},{"given":"Ralf","family":"Drechsel","sequence":"additional","affiliation":[{"name":"Process Optimization Ignition Systems - Paul Rauschert Steinbach GmbH Steinbach am Wald,Germany"}]},{"given":"Tobias","family":"Bocklet","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Department of Computer Science,Nurnberg,Germany"}]}],"member":"263","reference":[{"key":"ref10","author":"kingma","year":"2014","journal-title":"Adam A method for stochastic optimization"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-00296-0_5"},{"key":"ref12","article-title":"Evaluation: from precision, recall and f-measure to roc, informedness, markedness and correlation","author":"powers","year":"2020","journal-title":"ar Xiv preprint"},{"key":"ref13","author":"zhu","year":"2004","journal-title":"Recall Precision and Average Precision"},{"key":"ref14","article-title":"Attention is all you need","volume":"30","author":"vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.matt.2020.08.023"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1177\/1687814018755519"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/s10845-013-0761-y"},{"key":"ref5","doi-asserted-by":"crossref","first-page":"101538","DOI":"10.1016\/j.addma.2020.101538","article-title":"Machine learning in additive manufacturing: State-of-the-art and perspectives","volume":"36","author":"wang","year":"2020","journal-title":"Additive Manufacturing"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/0020-0190(87)90114-1"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1007\/s00170-019-03988-5"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1080\/21693277.2016.1192517"},{"key":"ref9","first-page":"8024","article-title":"Pytorch: An imperative style, high-performance deep learning library","author":"paszke","year":"2019","journal-title":"Advances in Neural IInformation Processing Systems"}],"event":{"name":"2022 IEEE Sensors","location":"Dallas, TX, USA","start":{"date-parts":[[2022,10,30]]},"end":{"date-parts":[[2022,11,2]]}},"container-title":["2022 IEEE Sensors"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9966996\/9966997\/09967124.pdf?arnumber=9967124","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,26]],"date-time":"2022-12-26T19:46:20Z","timestamp":1672083980000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9967124\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,30]]},"references-count":14,"URL":"http:\/\/dx.doi.org\/10.1109\/sensors52175.2022.9967124","relation":{},"subject":[],"published":{"date-parts":[[2022,10,30]]}}}</enrichment>
    <enrichment key="local_crossrefDocumentType">proceedings-article</enrichment>
    <enrichment key="local_crossrefLicence">https://doi.org/10.15223/policy-029</enrichment>
    <enrichment key="local_import_origin">crossref</enrichment>
    <enrichment key="local_doiImportPopulated">PersonAuthorFirstName_1,PersonAuthorLastName_1,PersonAuthorFirstName_2,PersonAuthorLastName_2,PersonAuthorFirstName_3,PersonAuthorLastName_3,PersonAuthorFirstName_4,PersonAuthorLastName_4,Enrichmentconference_title,Enrichmentconference_place,PublisherName,TitleMain_1,TitleParent_1,CompletedYear,Enrichmentlocal_crossrefLicence</enrichment>
    <enrichment key="conference_title">2022 IEEE Sensors</enrichment>
    <enrichment key="conference_place">Dallas, TX, USA</enrichment>
    <enrichment key="opus.source">doi-import</enrichment>
    <enrichment key="Reviewstatus">Begutachtet/Reviewed</enrichment>
    <enrichment key="ConferenceStatement">2022 IEEE Sensors, 30 October 2022 - 02 November 2022, Dallas, TX, USA</enrichment>
    <author>Thomas H. Schmitt</author>
    <author>Maximilian Bundscherer</author>
    <author>Ralf Drechsel</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Manufacturing</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>machine learning</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>optimization</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>ceramic bushing</value>
    </subject>
    <collection role="institutes" number="">Fakultät Informatik</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
  </doc>
  <doc>
    <id>2310</id>
    <completedYear>2024</completedYear>
    <publishedYear>2024</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>8</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>2024-06-30</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Machine Learning in Industrial Quality Control of Glass Bottle Prints</title>
    <abstract language="eng">In industrial manufacturing of glass bottles, quality control of bottle prints is necessary as numerous factors can negatively affect the printing process. Even minor defects in the bottle prints must be detected despite reflections in the glass or manufacturing-related deviations. In cooperation with our medium-sized industrial partner, two ML-based approaches for quality control of these bottle prints were developed and evaluated, which can also be used in this challenging scenario. Our first approach utilized different filters to supress reflections (e.g. Sobel or Canny) and image quality metrics for image comparison (e.g. MSE or SSIM) as features for different supervised classification models (e.g. SVM or k-Neighbors), which resulted in an accuracy of 84%. The images were aligned based on the ORB algorithm, which allowed us to estimate the rotations of the prints, which may serve as an indicator for anomalies in the manufacturing process. In our second approach, we fine-tuned different pre-trained CNN models (e.g. ResNet or VGG) for binary classification, which resulted in an accuracy of 87%. Utilizing Grad-Cam on our fine-tuned ResNet-34, we were able to localize and visualize frequently defective bottle print regions. This method allowed us to provide insights that could be used to optimize the actual manufacturing process. This paper also describes our general approach and the challenges we encountered in practice with data collection during ongoing production, unsupervised preselection, and labeling.</abstract>
    <identifier type="doi">10.48550/arXiv.2409.20132</identifier>
    <enrichment key="ConferenceStatement">KI in KMU 2024</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">false</enrichment>
    <licence>Creative Commons - CC BY - Namensnennung 4.0 International</licence>
    <author>Maximilian Bundscherer</author>
    <author>Thomas Schmitt</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Machine Learning, Quality Control, Industrial Manufacturing, Glass Bottle Printsuring Optimization, Glass Printing</value>
    </subject>
    <collection role="ddc" number="0">Informatik, Informationswissenschaft, allgemeine Werke</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
  </doc>
  <doc>
    <id>2307</id>
    <completedYear>2024</completedYear>
    <publishedYear>2024</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>6</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>report</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>2024-06-06</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Semmeldetector: Application of Machine Learning in Commercial Bakeries</title>
    <abstract language="eng">The Semmeldetector, is a machine learning application that utilizes object detection models to detect, classify and count baked goods in images. Our application allows commercial bakers to track unsold baked goods, which allows them to optimize production and increase resource efficiency. We compiled a dataset comprising 1151 images that distinguishes between 18 different types of baked goods to train our detection models. To facilitate model training, we used a Copy-Paste augmentation pipeline to expand our dataset. We trained the state-of-the-art object detection model YOLOv8 on our detection task. We tested the impact of different training data, model scale, and online image augmentation pipelines on model performance. Our overall best performing model, achieved an AP@0.5 of 89.1% on our test set. Based on our results, we conclude that machine learning can be a valuable tool even for unforeseen industries like bakeries, even with very limited datasets.</abstract>
    <identifier type="doi">10.48550/arXiv.2406.04050</identifier>
    <identifier type="arxiv">2406.04050</identifier>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">false</enrichment>
    <licence>Creative Commons - CC BY-NC-SA - Namensnennung - Nicht kommerziell -  Weitergabe unter gleichen Bedingungen 4.0 International</licence>
    <author>Thomas Schmitt</author>
    <author>Maximilian Bundscherer</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>machine learning, object detection, YOLOv8, image composition, baked goods, food inspection, industrial automation</value>
    </subject>
    <collection role="ddc" number="0">Informatik, Informationswissenschaft, allgemeine Werke</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
  </doc>
  <doc>
    <id>2308</id>
    <completedYear>2023</completedYear>
    <publishedYear>2023</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>14</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>article</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>2024-09-30</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Segmenting Wood Rot using Computer Vision Models</title>
    <abstract language="eng">In the woodworking industry, a huge amount of effort has to be invested into the initial quality assessment of the raw material. In this study we present an AI model to detect, quantify and localize defects on wooden logs. This model aims to both automate the quality control process and provide a more consistent and reliable quality assessment. For this purpose a dataset of 1424 sample images of wood logs is created. A total of 5 annotators possessing different levels of expertise is involved in dataset creation. An inter-annotator agreement analysis is conducted to analyze the impact of expertise on the annotation task and to highlight subjective differences in annotator judgement. We explore, train and fine-tune the state-of-the-art InternImage and ONE-PEACE architectures for semantic segmentation. The best model created achieves an average IoU of 0.71, and shows detection and quantification capabilities close to the human annotators.</abstract>
    <identifier type="arxiv">2409.20137v1</identifier>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">false</enrichment>
    <author>Roland Kammerbauer</author>
    <author>Thomas Schmitt</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>machine learning, image segmentation, semantic segmentation, InternImage, ONEPEACE, lumbering, industrial quality control, industrial automation</value>
    </subject>
    <collection role="ddc" number="0">Informatik, Informationswissenschaft, allgemeine Werke</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
  </doc>
  <doc>
    <id>2309</id>
    <completedYear>2024</completedYear>
    <publishedYear>2024</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>10</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>article</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>2024-09-30</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Training a Computer Vision Model for Commercial Bakeries with Primarily Synthetic Images</title>
    <abstract language="eng">In the food industry, reprocessing returned product is a vital step to increase resource efficiency. [SBB23] presented an AI application that automates the tracking of returned bread buns. We extend their work by creating an expanded dataset comprising 2432 images and a wider range of baked goods. To increase model robustness, we use generative models pix2pix and CycleGAN to create synthetic images. We train state-of-the-art object detection model YOLOv9 and YOLOv8 on our detection task. Our overall best-performing model achieved an average precision AP@0.5 of 90.3% on our test set.</abstract>
    <identifier type="doi">10.48550/arXiv.2409.20122</identifier>
    <identifier type="arxiv">2409.20122v1</identifier>
    <enrichment key="ConferenceStatement">FZI Workshop - Künstliche Intelligenz im Mittelstand (KI-KMU 2024)</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">false</enrichment>
    <licence>Creative Commons - CC BY-NC-SA - Namensnennung - Nicht kommerziell -  Weitergabe unter gleichen Bedingungen 4.0 International</licence>
    <author>Thomas Schmitt</author>
    <author>Maximilian Bundscherer</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>machine learning, object detection, YOLOv9, image composition, baked goods, food industry, industrial automation</value>
    </subject>
    <collection role="ddc" number="0">Informatik, Informationswissenschaft, allgemeine Werke</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
  </doc>
  <doc>
    <id>4350</id>
    <completedYear>2025</completedYear>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1</pageFirst>
    <pageLast>5</pageLast>
    <pageNumber>5</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>2025-01-13</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Digital Operating Mode Classification of Real-World Amateur Radio Transmissions</title>
    <abstract language="eng">This study presents an ML approach for classifying digital radio operating modes evaluated on real-world transmissions. We generated 98 different parameterized radio signals from 17 digital operating modes, transmitted each of them on the 70 cm (UHF) amateur radio band, and recorded our transmissions with two different architectures of SDR receivers. Three lightweight ML models were trained exclusively on spectrograms of limited non-transmitted signals with random characters as payloads. This training involved an online data augmentation pipeline to simulate various radio channel impairments. Our best model, EfficientNetB0, achieved an accuracy of 93.80% across the 17 operating modes and 85.47% across all 98 parameterized radio signals, evaluated on our real-world transmissions with&#13;
Wikipedia articles as payloads. Furthermore, we analyzed the impact of varying signal durations &amp; the number of FFT bins on classification, assessed the effectiveness of our simulated channel impairments, and tested our models across multiple simulated SNRs.</abstract>
    <parentTitle language="eng">ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</parentTitle>
    <identifier type="doi">10.1109/ICASSP49660.2025.10889837</identifier>
    <identifier type="arxiv">2501.07337</identifier>
    <enrichment key="opus_doi_flag">true</enrichment>
    <enrichment key="opus_import_data">{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,2]],"date-time":"2025-08-02T18:43:58Z","timestamp":1754160238273,"version":"3.41.2"},"reference-count":32,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100004326","name":"Bayer","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004326","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,4,6]]},"DOI":"10.1109\/icassp49660.2025.10889837","type":"proceedings-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T13:52:43Z","timestamp":1741787563000},"page":"1-5","source":"Crossref","is-referenced-by-count":0,"title":["Digital Operating Mode Classification of Real-World Amateur Radio Transmissions"],"prefix":"10.1109","author":[{"given":"Maximilian","family":"Bundscherer","sequence":"first","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Nuremberg,Germany"}]},{"given":"Thomas","family":"Schmitt","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Nuremberg,Germany"}]},{"given":"Ilja","family":"Baumann","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Nuremberg,Germany"}]},{"given":"Tobias","family":"Bocklet","sequence":"additional","affiliation":[{"name":"Technische Hochschule N&amp;#x00FC;rnberg Georg Simon Ohm,Nuremberg,Germany"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICEENG45378.2020.9171706"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/LCOMM.2020.2968030"},{"article-title":"Large scale radio frequency signal classification","year":"2022","author":"Boegner","key":"ref3"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/WCSP49889.2020.9299821"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/WCSP49889.2020.9299797"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TVT.2018.2868698"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2960775"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/WOCC.2017.7929000"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2021.3091523"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TCCN.2020.3023145"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/LWC.2022.3217292"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TCCN.2021.3120997"},{"article-title":"Exact signal measurements using fft analysis","year":"2016","author":"Scholl","key":"ref13"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ISdea.2012.487"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/VTC2021-Spring51267.2021.9449073"},{"key":"ref16","doi-asserted-by":"crossref","DOI":"10.31219\/osf.io\/cvus7","article-title":"Large scale radio frequency wideband signal detection &amp; recognition","author":"Boegner","year":"2022"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/WOCC58016.2023.10139594"},{"issue":"1","key":"ref18","article-title":"Radio machine learning dataset generation with gnu radio","volume-title":"Proceedings of the GNU radio conference","volume":"1","author":"O\u2019shea"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2018.2797022"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1016\/j.cja.2021.08.016"},{"volume-title":"Wireless Communications - Principles and Practice","year":"2002","author":"Rappaport","key":"ref21"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511807213"},{"volume-title":"Understanding Digital Signal Processing","year":"2010","author":"Lyons","key":"ref23"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-662-53234-8"},{"volume-title":"Software Radio - A Modern Approach to Radio Engineering","year":"2002","author":"Reed","key":"ref25"},{"volume-title":"Digital Communications","year":"2008","author":"Proakis","key":"ref26"},{"volume-title":"ARRL\u2019s HF Digital Handbook","year":"2007","author":"Ford","key":"ref27"},{"article-title":"Fldigi","year":"2023","author":"Freese","key":"ref28"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref30","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume-title":"Proceedings of the 36th International Conference on Machine Learning, ser. Proceedings of Machine Learning Research","volume":"97","author":"Tan"},{"article-title":"Vision mamba: Efficient visual representation learning with bidirectional state space model","year":"2024","author":"Zhu","key":"ref31"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"}],"event":{"name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2025,4,6]]},"location":"Hyderabad, India","end":{"date-parts":[[2025,4,11]]}},"container-title":["ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10887540\/10887541\/10889837.pdf?arnumber=10889837","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,26]],"date-time":"2025-07-26T06:55:42Z","timestamp":1753512942000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10889837\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,6]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/icassp49660.2025.10889837","relation":{},"subject":[],"published":{"date-parts":[[2025,4,6]]}}}</enrichment>
    <enrichment key="local_crossrefDocumentType">proceedings-article</enrichment>
    <enrichment key="local_crossrefLicence">https://doi.org/10.15223/policy-029</enrichment>
    <enrichment key="local_import_origin">crossref</enrichment>
    <enrichment key="local_doiImportPopulated">PersonAuthorFirstName_1,PersonAuthorLastName_1,PersonAuthorFirstName_2,PersonAuthorLastName_2,PersonAuthorFirstName_3,PersonAuthorLastName_3,PersonAuthorFirstName_4,PersonAuthorLastName_4,Enrichmentconference_title,Enrichmentconference_place,PublisherName,TitleMain_1,TitleParent_1,PageNumber,PageFirst,PageLast,CompletedYear,Enrichmentlocal_crossrefLicence</enrichment>
    <enrichment key="conference_title">ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</enrichment>
    <enrichment key="conference_place">Hyderabad, India</enrichment>
    <enrichment key="opus.source">doi-import</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">false</enrichment>
    <licence>Creative Commons - CC BY - Namensnennung 4.0 International</licence>
    <author>Maximilian Bundscherer</author>
    <author>Thomas Schmitt</author>
    <author>Ilja Baumann</author>
    <author>Tobias Bocklet</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Automatic Modulation Classification</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Amateur</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Radio</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Spectrum Monitoring</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Cognitive Radio</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Machine Learning</value>
    </subject>
    <collection role="institutes" number="">Fakultät Informatik</collection>
    <collection role="Forschungsschwerpunkt" number="5">Digitalisierung &amp; Künstliche Intelligenz</collection>
    <collection role="institutes" number="">Zentrum für Künstliche Intelligenz (KIZ)</collection>
    <file>https://opus4.kobv.de/opus4-ohm/files/4350/ICASSP25_bundscherer_digiModeClass_accepted.pdf</file>
  </doc>
</export-example>
