{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etdr-1611"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etdr-1611","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"BRUISE DETECTION IN APPLES USING 3D INFRARED IMAGING AND MACHINE LEARNING TECHNOLOGIES","abstract":"Bruise detection plays an important role in fruit grading. A bruise detection system capable of finding and removing damaged products on the production lines will distinctly improve the quality of fruits for sale, and consequently improve the fruit economy. This dissertation presents a novel automatic detection system based on surface information obtained from 3D near-infrared imaging technique for bruised apple identification. The proposed 3D bruise detection system is expected to provide better performance in bruise detection than the existing 2D systems. We first propose a mesh denoising filter to reduce noise effect while preserving the geometric features of the meshes. Compared with several existing mesh denoising filters, the proposed filter achieves better performance in reducing noise effect as well as preserving bruised regions in 3D meshes of bruised apples. Next, we investigate two different machine learning techniques for the identification of bruised apples. The first technique is to extract hand-crafted feature from 3D meshes, and train a predictive classifier based on hand-crafted features. It is shown that the predictive model trained on the proposed hand-crafted features outperforms the same models trained on several other local shape descriptors. The second technique is to apply deep learning to learn the feature representation automatically from the mesh data, and then use the deep learning model or a new predictive model for the classification. The optimized deep learning model achieves very high classification accuracy, and it outperforms the performance of the detection system based on the proposed hand-crafted features. At last, we investigate GPU techniques for accelerating the proposed apple bruise detection system. Specifically, the dissertation proposes a GPU framework, implemented in CUDA, for the acceleration of the algorithm that extracts vertex-based local binary patterns. Experimental results show that the proposed GPU program speeds up the process of extracting local binary patterns by 5 times compared to a single-core CPU program.","abstract_html":"Bruise detection plays an important role in fruit grading. A bruise detection system capable of finding and removing damaged products on the production lines will distinctly improve the quality of fruits for sale, and consequently improve the fruit economy. This dissertation presents a novel automatic detection system based on surface information obtained from 3D near-infrared imaging technique for bruised apple identification. The proposed 3D bruise detection system is expected to provide better performance in bruise detection than the existing 2D systems. We first propose a mesh denoising filter to reduce noise effect while preserving the geometric features of the meshes. Compared with several existing mesh denoising filters, the proposed filter achieves better performance in reducing noise effect as well as preserving bruised regions in 3D meshes of bruised apples. Next, we investigate two different machine learning techniques for the identification of bruised apples. The first technique is to extract hand-crafted feature from 3D meshes, and train a predictive classifier based on hand-crafted features. It is shown that the predictive model trained on the proposed hand-crafted features outperforms the same models trained on several other local shape descriptors. The second technique is to apply deep learning to learn the feature representation automatically from the mesh data, and then use the deep learning model or a new predictive model for the classification. The optimized deep learning model achieves very high classification accuracy, and it outperforms the performance of the detection system based on the proposed hand-crafted features. At last, we investigate GPU techniques for accelerating the proposed apple bruise detection system. Specifically, the dissertation proposes a GPU framework, implemented in CUDA, for the acceleration of the algorithm that extracts vertex-based local binary patterns. Experimental results show that the proposed GPU program speeds up the process of extracting local binary patterns by 5 times compared to a single-core CPU program.","abstract_has_math":false,"creators":["Hu, Zilong"],"institution":null,"degree_name":"Doctor of Philosophy in Computational Science and Engineering (PhD)","degree_level":"Open Access Dissertation","degree_discipline":"School of Technology","degree_department":null,"school":null,"contributors":["Jinshan Tang"],"advisors":[],"committee_chairs":[],"committee_members":["Guiliang Tang","Yu Cai","Eugene Levin"],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:17:09Z","subjects":["Fruit bruise detection","3D near-infrared imaging","machine learning","Computational Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etdr/509"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etdr/509","href":"https://digitalcommons.mtu.edu/etdr/509","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etdr/509","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etdr/509","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.mtu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.mtu.edu%3Aetdr-1611","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Jinshan Tang"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Guiliang Tang","Yu Cai","Eugene Levin"]},{"key":"dc:creator","label":"Author","values":["Hu, Zilong"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2018-12-04T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2017-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["School of Technology"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Open Access Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy in Computational Science and Engineering (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Fruit bruise detection","3D near-infrared imaging","machine learning","Computational Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etdr/509","https://digitalcommons.mtu.edu/etdr/509"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Bruise detection plays an important role in fruit grading. A bruise detection system capable of finding and removing damaged products on the production lines will distinctly improve the quality of fruits for sale, and consequently improve the fruit economy. This dissertation presents a novel automatic detection system based on surface information obtained from 3D near-infrared imaging technique for bruised apple identification. The proposed 3D bruise detection system is expected to provide better performance in bruise detection than the existing 2D systems. We first propose a mesh denoising filter to reduce noise effect while preserving the geometric features of the meshes. Compared with several existing mesh denoising filters, the proposed filter achieves better performance in reducing noise effect as well as preserving bruised regions in 3D meshes of bruised apples. Next, we investigate two different machine learning techniques for the identification of bruised apples. The first technique is to extract hand-crafted feature from 3D meshes, and train a predictive classifier based on hand-crafted features. It is shown that the predictive model trained on the proposed hand-crafted features outperforms the same models trained on several other local shape descriptors. The second technique is to apply deep learning to learn the feature representation automatically from the mesh data, and then use the deep learning model or a new predictive model for the classification. The optimized deep learning model achieves very high classification accuracy, and it outperforms the performance of the detection system based on the proposed hand-crafted features. At last, we investigate GPU techniques for accelerating the proposed apple bruise detection system. Specifically, the dissertation proposes a GPU framework, implemented in CUDA, for the acceleration of the algorithm that extracts vertex-based local binary patterns. Experimental results show that the proposed GPU program speeds up the process of extracting local binary patterns by 5 times compared to a single-core CPU program."]},{"key":"dc:title","label":"Title","values":["BRUISE DETECTION IN APPLES USING 3D INFRARED IMAGING AND MACHINE LEARNING TECHNOLOGIES"]}]}],"canonical_facts":{"dc:contributor":["Jinshan Tang"],"dc:contributor.committeemember":["Guiliang Tang","Yu Cai","Eugene Levin"],"dc:creator":["Hu, Zilong"],"dc:date.available":["2018-12-04T08:00:00Z"],"dc:date.created":["2017-01-01T08:00:00Z"],"dc:description.abstract":["Bruise detection plays an important role in fruit grading. A bruise detection system capable of finding and removing damaged products on the production lines will distinctly improve the quality of fruits for sale, and consequently improve the fruit economy. This dissertation presents a novel automatic detection system based on surface information obtained from 3D near-infrared imaging technique for bruised apple identification. The proposed 3D bruise detection system is expected to provide better performance in bruise detection than the existing 2D systems. We first propose a mesh denoising filter to reduce noise effect while preserving the geometric features of the meshes. Compared with several existing mesh denoising filters, the proposed filter achieves better performance in reducing noise effect as well as preserving bruised regions in 3D meshes of bruised apples. Next, we investigate two different machine learning techniques for the identification of bruised apples. The first technique is to extract hand-crafted feature from 3D meshes, and train a predictive classifier based on hand-crafted features. It is shown that the predictive model trained on the proposed hand-crafted features outperforms the same models trained on several other local shape descriptors. The second technique is to apply deep learning to learn the feature representation automatically from the mesh data, and then use the deep learning model or a new predictive model for the classification. The optimized deep learning model achieves very high classification accuracy, and it outperforms the performance of the detection system based on the proposed hand-crafted features. At last, we investigate GPU techniques for accelerating the proposed apple bruise detection system. Specifically, the dissertation proposes a GPU framework, implemented in CUDA, for the acceleration of the algorithm that extracts vertex-based local binary patterns. Experimental results show that the proposed GPU program speeds up the process of extracting local binary patterns by 5 times compared to a single-core CPU program."],"dc:identifier":["10.37099/mtu.dc.etdr/509","https://digitalcommons.mtu.edu/etdr/509"],"dc:subject":["Fruit bruise detection","3D near-infrared imaging","machine learning","Computational Engineering"],"dc:title":["BRUISE DETECTION IN APPLES USING 3D INFRARED IMAGING AND MACHINE LEARNING TECHNOLOGIES"],"thesis:degree_discipline":["School of Technology"],"thesis:degree_level":["Open Access Dissertation"],"thesis:degree_name":["Doctor of Philosophy in Computational Science and Engineering (PhD)"]},"updated_at":"2026-11-19T08:17:09Z"}