Transfer learning for galaxy feature detection: Finding Giant Star-forming Clumps in low redshift galaxies using Faster R-CNN
Giant Star-forming Clumps (GSFCs) are areas of intensive star-formation that are commonly observed in high-redshift (z>1) galaxies but their formation and role in galaxy evolution remain unclear. High-resolution observations of low-redshift clumpy galaxy analogues are rare and restricted to a lim...
Gespeichert in:
| Veröffentlicht in: | arXiv.org 2024-04 |
|---|---|
| Hauptverfasser: | , , , , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | arXiv.org |
| container_volume | |
| creator | Popp, Jürgen Dickinson, Hugh Serjeant, Stephen Walmsley, Mike Adams, Dominic tson, Lucy Mantha, Kameswara Mehta, Vihang Dawson, James M Kruk, Sandor Simmons, Brooke |
| description | Giant Star-forming Clumps (GSFCs) are areas of intensive star-formation that are commonly observed in high-redshift (z>1) galaxies but their formation and role in galaxy evolution remain unclear. High-resolution observations of low-redshift clumpy galaxy analogues are rare and restricted to a limited set of galaxies but the increasing availability of wide-field galaxy survey data makes the detection of large clumpy galaxy samples increasingly feasible. Deep Learning, and in particular CNNs, have been successfully applied to image classification tasks in astrophysical data analysis. However, one application of DL that remains relatively unexplored is that of automatically identifying and localising specific objects or features in astrophysical imaging data. In this paper we demonstrate the feasibility of using Deep learning-based object detection models to localise GSFCs in astrophysical imaging data. We apply the Faster R-CNN object detection framework (FRCNN) to identify GSFCs in low redshift (z |
| format | Article |
| fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2899302948</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2899302948</sourcerecordid><originalsourceid>FETCH-proquest_journals_28993029483</originalsourceid><addsrcrecordid>eNqNjcFuwjAQRC2kSiDKP6zEOVKwoSRcI0JPHFruaNWswcjYsGur5dB_b6L2A3oaaebNzEhNtDGLolpqPVYzkUtZlvplrVcrM1HfB8Yglhg8IQcXTmAjwwk9fj3AEqbMBB0l-kguhg20LnQDtXMYErwn5KIvXAer8fl6E3ABfPwEpk7OzqbfLUcCWQaqRUn93VvR7PfP6smiF5r96VTN2-2heS1uHO-ZJB0vMXPoo6Ou6tqUul5W5n_UDw9kT00</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2899302948</pqid></control><display><type>article</type><title>Transfer learning for galaxy feature detection: Finding Giant Star-forming Clumps in low redshift galaxies using Faster R-CNN</title><source>Free E- Journals</source><creator>Popp, Jürgen ; Dickinson, Hugh ; Serjeant, Stephen ; Walmsley, Mike ; Adams, Dominic ; tson, Lucy ; Mantha, Kameswara ; Mehta, Vihang ; Dawson, James M ; Kruk, Sandor ; Simmons, Brooke</creator><creatorcontrib>Popp, Jürgen ; Dickinson, Hugh ; Serjeant, Stephen ; Walmsley, Mike ; Adams, Dominic ; tson, Lucy ; Mantha, Kameswara ; Mehta, Vihang ; Dawson, James M ; Kruk, Sandor ; Simmons, Brooke</creatorcontrib><description>Giant Star-forming Clumps (GSFCs) are areas of intensive star-formation that are commonly observed in high-redshift (z>1) galaxies but their formation and role in galaxy evolution remain unclear. High-resolution observations of low-redshift clumpy galaxy analogues are rare and restricted to a limited set of galaxies but the increasing availability of wide-field galaxy survey data makes the detection of large clumpy galaxy samples increasingly feasible. Deep Learning, and in particular CNNs, have been successfully applied to image classification tasks in astrophysical data analysis. However, one application of DL that remains relatively unexplored is that of automatically identifying and localising specific objects or features in astrophysical imaging data. In this paper we demonstrate the feasibility of using Deep learning-based object detection models to localise GSFCs in astrophysical imaging data. We apply the Faster R-CNN object detection framework (FRCNN) to identify GSFCs in low redshift (z<0.3) galaxies. Unlike other studies, we train different FRCNN models not on simulated images with known labels but on real observational data that was collected by the Sloan Digital Sky Survey Legacy Survey and labelled by volunteers from the citizen science project `Galaxy Zoo: Clump Scout'. The FRCNN model relies on a CNN component as a `backbone' feature extractor. We show that CNNs, that have been pre-trained for image classification using astrophysical images, outperform those that have been pre-trained on terrestrial images. In particular, we compare a domain-specific CNN -`Zoobot' - with a generic classification backbone and find that Zoobot achieves higher detection performance and also requires smaller training data sets to do so. Our final model is capable of producing GSFC detections with a completeness and purity of >=0.8 while only being trained on ~5,000 galaxy images.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Astronomical models ; Clumps ; Data analysis ; Deep learning ; Feasibility ; Galactic evolution ; Galaxies ; Giant stars ; Image classification ; Machine learning ; Object recognition ; Red shift ; Sky surveys (astronomy) ; Star formation ; Stars & galaxies</subject><ispartof>arXiv.org, 2024-04</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Popp, Jürgen</creatorcontrib><creatorcontrib>Dickinson, Hugh</creatorcontrib><creatorcontrib>Serjeant, Stephen</creatorcontrib><creatorcontrib>Walmsley, Mike</creatorcontrib><creatorcontrib>Adams, Dominic</creatorcontrib><creatorcontrib>tson, Lucy</creatorcontrib><creatorcontrib>Mantha, Kameswara</creatorcontrib><creatorcontrib>Mehta, Vihang</creatorcontrib><creatorcontrib>Dawson, James M</creatorcontrib><creatorcontrib>Kruk, Sandor</creatorcontrib><creatorcontrib>Simmons, Brooke</creatorcontrib><title>Transfer learning for galaxy feature detection: Finding Giant Star-forming Clumps in low redshift galaxies using Faster R-CNN</title><title>arXiv.org</title><description>Giant Star-forming Clumps (GSFCs) are areas of intensive star-formation that are commonly observed in high-redshift (z>1) galaxies but their formation and role in galaxy evolution remain unclear. High-resolution observations of low-redshift clumpy galaxy analogues are rare and restricted to a limited set of galaxies but the increasing availability of wide-field galaxy survey data makes the detection of large clumpy galaxy samples increasingly feasible. Deep Learning, and in particular CNNs, have been successfully applied to image classification tasks in astrophysical data analysis. However, one application of DL that remains relatively unexplored is that of automatically identifying and localising specific objects or features in astrophysical imaging data. In this paper we demonstrate the feasibility of using Deep learning-based object detection models to localise GSFCs in astrophysical imaging data. We apply the Faster R-CNN object detection framework (FRCNN) to identify GSFCs in low redshift (z<0.3) galaxies. Unlike other studies, we train different FRCNN models not on simulated images with known labels but on real observational data that was collected by the Sloan Digital Sky Survey Legacy Survey and labelled by volunteers from the citizen science project `Galaxy Zoo: Clump Scout'. The FRCNN model relies on a CNN component as a `backbone' feature extractor. We show that CNNs, that have been pre-trained for image classification using astrophysical images, outperform those that have been pre-trained on terrestrial images. In particular, we compare a domain-specific CNN -`Zoobot' - with a generic classification backbone and find that Zoobot achieves higher detection performance and also requires smaller training data sets to do so. Our final model is capable of producing GSFC detections with a completeness and purity of >=0.8 while only being trained on ~5,000 galaxy images.</description><subject>Astronomical models</subject><subject>Clumps</subject><subject>Data analysis</subject><subject>Deep learning</subject><subject>Feasibility</subject><subject>Galactic evolution</subject><subject>Galaxies</subject><subject>Giant stars</subject><subject>Image classification</subject><subject>Machine learning</subject><subject>Object recognition</subject><subject>Red shift</subject><subject>Sky surveys (astronomy)</subject><subject>Star formation</subject><subject>Stars & galaxies</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNjcFuwjAQRC2kSiDKP6zEOVKwoSRcI0JPHFruaNWswcjYsGur5dB_b6L2A3oaaebNzEhNtDGLolpqPVYzkUtZlvplrVcrM1HfB8Yglhg8IQcXTmAjwwk9fj3AEqbMBB0l-kguhg20LnQDtXMYErwn5KIvXAer8fl6E3ABfPwEpk7OzqbfLUcCWQaqRUn93VvR7PfP6smiF5r96VTN2-2heS1uHO-ZJB0vMXPoo6Ou6tqUul5W5n_UDw9kT00</recordid><startdate>20240401</startdate><enddate>20240401</enddate><creator>Popp, Jürgen</creator><creator>Dickinson, Hugh</creator><creator>Serjeant, Stephen</creator><creator>Walmsley, Mike</creator><creator>Adams, Dominic</creator><creator>tson, Lucy</creator><creator>Mantha, Kameswara</creator><creator>Mehta, Vihang</creator><creator>Dawson, James M</creator><creator>Kruk, Sandor</creator><creator>Simmons, Brooke</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20240401</creationdate><title>Transfer learning for galaxy feature detection: Finding Giant Star-forming Clumps in low redshift galaxies using Faster R-CNN</title><author>Popp, Jürgen ; Dickinson, Hugh ; Serjeant, Stephen ; Walmsley, Mike ; Adams, Dominic ; tson, Lucy ; Mantha, Kameswara ; Mehta, Vihang ; Dawson, James M ; Kruk, Sandor ; Simmons, Brooke</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28993029483</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Astronomical models</topic><topic>Clumps</topic><topic>Data analysis</topic><topic>Deep learning</topic><topic>Feasibility</topic><topic>Galactic evolution</topic><topic>Galaxies</topic><topic>Giant stars</topic><topic>Image classification</topic><topic>Machine learning</topic><topic>Object recognition</topic><topic>Red shift</topic><topic>Sky surveys (astronomy)</topic><topic>Star formation</topic><topic>Stars & galaxies</topic><toplevel>online_resources</toplevel><creatorcontrib>Popp, Jürgen</creatorcontrib><creatorcontrib>Dickinson, Hugh</creatorcontrib><creatorcontrib>Serjeant, Stephen</creatorcontrib><creatorcontrib>Walmsley, Mike</creatorcontrib><creatorcontrib>Adams, Dominic</creatorcontrib><creatorcontrib>tson, Lucy</creatorcontrib><creatorcontrib>Mantha, Kameswara</creatorcontrib><creatorcontrib>Mehta, Vihang</creatorcontrib><creatorcontrib>Dawson, James M</creatorcontrib><creatorcontrib>Kruk, Sandor</creatorcontrib><creatorcontrib>Simmons, Brooke</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection (ProQuest)</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Popp, Jürgen</au><au>Dickinson, Hugh</au><au>Serjeant, Stephen</au><au>Walmsley, Mike</au><au>Adams, Dominic</au><au>tson, Lucy</au><au>Mantha, Kameswara</au><au>Mehta, Vihang</au><au>Dawson, James M</au><au>Kruk, Sandor</au><au>Simmons, Brooke</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Transfer learning for galaxy feature detection: Finding Giant Star-forming Clumps in low redshift galaxies using Faster R-CNN</atitle><jtitle>arXiv.org</jtitle><date>2024-04-01</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Giant Star-forming Clumps (GSFCs) are areas of intensive star-formation that are commonly observed in high-redshift (z>1) galaxies but their formation and role in galaxy evolution remain unclear. High-resolution observations of low-redshift clumpy galaxy analogues are rare and restricted to a limited set of galaxies but the increasing availability of wide-field galaxy survey data makes the detection of large clumpy galaxy samples increasingly feasible. Deep Learning, and in particular CNNs, have been successfully applied to image classification tasks in astrophysical data analysis. However, one application of DL that remains relatively unexplored is that of automatically identifying and localising specific objects or features in astrophysical imaging data. In this paper we demonstrate the feasibility of using Deep learning-based object detection models to localise GSFCs in astrophysical imaging data. We apply the Faster R-CNN object detection framework (FRCNN) to identify GSFCs in low redshift (z<0.3) galaxies. Unlike other studies, we train different FRCNN models not on simulated images with known labels but on real observational data that was collected by the Sloan Digital Sky Survey Legacy Survey and labelled by volunteers from the citizen science project `Galaxy Zoo: Clump Scout'. The FRCNN model relies on a CNN component as a `backbone' feature extractor. We show that CNNs, that have been pre-trained for image classification using astrophysical images, outperform those that have been pre-trained on terrestrial images. In particular, we compare a domain-specific CNN -`Zoobot' - with a generic classification backbone and find that Zoobot achieves higher detection performance and also requires smaller training data sets to do so. Our final model is capable of producing GSFC detections with a completeness and purity of >=0.8 while only being trained on ~5,000 galaxy images.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | EISSN: 2331-8422 |
| ispartof | arXiv.org, 2024-04 |
| issn | 2331-8422 |
| language | eng |
| recordid | cdi_proquest_journals_2899302948 |
| source | Free E- Journals |
| subjects | Astronomical models Clumps Data analysis Deep learning Feasibility Galactic evolution Galaxies Giant stars Image classification Machine learning Object recognition Red shift Sky surveys (astronomy) Star formation Stars & galaxies |
| title | Transfer learning for galaxy feature detection: Finding Giant Star-forming Clumps in low redshift galaxies using Faster R-CNN |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-24T10%3A43%3A47IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Transfer%20learning%20for%20galaxy%20feature%20detection:%20Finding%20Giant%20Star-forming%20Clumps%20in%20low%20redshift%20galaxies%20using%20Faster%20R-CNN&rft.jtitle=arXiv.org&rft.au=Popp,%20J%C3%BCrgen&rft.date=2024-04-01&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2899302948%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2899302948&rft_id=info:pmid/&rfr_iscdi=true |