Loading…
trackdem: Automated particle tracking to obtain population counts and size distributions from videos in r
The possibilities for image analysis in scientific research are substantial: the costs of digital cameras and data storage are sharply decreasing, and automated image analyses greatly increase the scale, reproducibility and robustness of biological studies. However, automated image analysis in ecolo...
Saved in:
| Published in: | Methods in ecology and evolution 2018-04, Vol.9 (4), p.965-973 |
|---|---|
| Main Authors: | , , , , |
| Format: | Article |
| Language: | English |
| Subjects: | |
| Citations: | Items that this one cites Items that cite this one |
| Online Access: | Request full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| cited_by | cdi_FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830 |
|---|---|
| cites | cdi_FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830 |
| container_end_page | 973 |
| container_issue | 4 |
| container_start_page | 965 |
| container_title | Methods in ecology and evolution |
| container_volume | 9 |
| creator | Bruijning, Marjolein Visser, Marco D. Hallmann, Caspar A. Jongejans, Eelke Golding, Nick |
| description | The possibilities for image analysis in scientific research are substantial: the costs of digital cameras and data storage are sharply decreasing, and automated image analyses greatly increase the scale, reproducibility and robustness of biological studies. However, automated image analysis in ecological and evolutionary studies is still in its infancy. There is a clear need for easy to use and accessible tools.
Here, we provide a general purpose method to obtain estimates of population densities, individual body sizes and behavioural metrics from video material of moving organisms. The methods are supplied as a new r‐package trackdem, which provides a flexible, easy to install and use, generally applicable and accurate way to analyse ecological video data. The package can detect and track moving particles, count individuals and estimate individual sizes using background detection, particle identification and particle tracking algorithms. Machine learning is implemented to reduce the influence of noise in lower quality videos or to distinguish a single species in multi‐species systems.
We show that trackdem provides accurate population counts and body size distributions. Using a series of simulations, we show that our estimates are robust against high levels of noise in videos. When applied to live populations of Daphnia magna, our methods obtained accurate and unbiased estimates of population counts, individual sizes and size distributions, as verified by manual counting and measuring. The package trackdem is also directly usable for movement analysis, for instance in behavioural ecology, as illustrated by the tracking of insects, fish, cars and humans.
Within 24 hr, we obtained 192 accurate population counts and body sizes of 22,154 individuals. Such results underscore that automated analysis can improve robustness and reproducibility, and greatly increase the scope of studies in ecology and evolution. |
| doi_str_mv | 10.1111/2041-210X.12975 |
| format | article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2022940899</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2022940899</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830</originalsourceid><addsrcrecordid>eNqFkMtLxDAQxoMouKx79hrw3N08-ki9Lcv6gBUvCt5CmodkbZuapMr619tuRbw5lxnm-34z8AFwidESD7UiKMUJwehliUlZZCdg9rs5_TOfg0UIezQUZSUi6QzY6IV8U7q5hus-ukZErWAnfLSy1vAo2vYVRgddFYVtYee6vhbRuhZK17cxQNEqGOyXhsqG6G3Vj2KAxrsGflilXYAD5y_AmRF10IufPgfPN9unzV2ye7y936x3iaRZkSWaGlFoZjQmBUlNJbKM0hxpygzJcp2rSkqM0rSiuSoIyytkRC4VopKxgjKK5uBqutt5997rEPne9b4dXnKCCClTxMpycK0ml_QuBK8N77xthD9wjPgYKR9D42No_BjpQOQT8WlrffjPzh-2WzqB30VDeZ4</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2022940899</pqid></control><display><type>article</type><title>trackdem: Automated particle tracking to obtain population counts and size distributions from videos in r</title><creator>Bruijning, Marjolein ; Visser, Marco D. ; Hallmann, Caspar A. ; Jongejans, Eelke ; Golding, Nick</creator><contributor>Golding, Nick</contributor><creatorcontrib>Bruijning, Marjolein ; Visser, Marco D. ; Hallmann, Caspar A. ; Jongejans, Eelke ; Golding, Nick ; Golding, Nick</creatorcontrib><description>The possibilities for image analysis in scientific research are substantial: the costs of digital cameras and data storage are sharply decreasing, and automated image analyses greatly increase the scale, reproducibility and robustness of biological studies. However, automated image analysis in ecological and evolutionary studies is still in its infancy. There is a clear need for easy to use and accessible tools.
Here, we provide a general purpose method to obtain estimates of population densities, individual body sizes and behavioural metrics from video material of moving organisms. The methods are supplied as a new r‐package trackdem, which provides a flexible, easy to install and use, generally applicable and accurate way to analyse ecological video data. The package can detect and track moving particles, count individuals and estimate individual sizes using background detection, particle identification and particle tracking algorithms. Machine learning is implemented to reduce the influence of noise in lower quality videos or to distinguish a single species in multi‐species systems.
We show that trackdem provides accurate population counts and body size distributions. Using a series of simulations, we show that our estimates are robust against high levels of noise in videos. When applied to live populations of Daphnia magna, our methods obtained accurate and unbiased estimates of population counts, individual sizes and size distributions, as verified by manual counting and measuring. The package trackdem is also directly usable for movement analysis, for instance in behavioural ecology, as illustrated by the tracking of insects, fish, cars and humans.
Within 24 hr, we obtained 192 accurate population counts and body sizes of 22,154 individuals. Such results underscore that automated analysis can improve robustness and reproducibility, and greatly increase the scope of studies in ecology and evolution.</description><identifier>ISSN: 2041-210X</identifier><identifier>EISSN: 2041-210X</identifier><identifier>DOI: 10.1111/2041-210X.12975</identifier><language>eng</language><publisher>London: John Wiley & Sons, Inc</publisher><subject>automated population counts ; Automation ; Biological evolution ; Body size ; Cameras ; Computer simulation ; Cost analysis ; Counting ; Data processing ; Data storage ; Digital cameras ; Digital imaging ; Ecological monitoring ; Ecology ; Estimates ; Image analysis ; Image processing ; individual trajectories ; Insects ; Learning algorithms ; Machine learning ; movement behaviour ; neural net ; noise filtering ; Noise reduction ; particle identification ; Particle tracking ; Population ; Population statistics ; Reproducibility ; Robustness ; size distribution ; Video data</subject><ispartof>Methods in ecology and evolution, 2018-04, Vol.9 (4), p.965-973</ispartof><rights>2018 The Authors. Methods in Ecology and Evolution © 2018 British Ecological Society</rights><rights>Methods in Ecology and Evolution © 2018 British Ecological Society</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830</citedby><cites>FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830</cites><orcidid>0000-0002-4630-0522 ; 0000-0002-9408-2187 ; 0000-0003-1200-0852 ; 0000-0003-1148-7419</orcidid></display><links><openurl>$$Topenurl_article</openurl><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>783</link.rule.ids></links><search><contributor>Golding, Nick</contributor><creatorcontrib>Bruijning, Marjolein</creatorcontrib><creatorcontrib>Visser, Marco D.</creatorcontrib><creatorcontrib>Hallmann, Caspar A.</creatorcontrib><creatorcontrib>Jongejans, Eelke</creatorcontrib><creatorcontrib>Golding, Nick</creatorcontrib><title>trackdem: Automated particle tracking to obtain population counts and size distributions from videos in r</title><title>Methods in ecology and evolution</title><description>The possibilities for image analysis in scientific research are substantial: the costs of digital cameras and data storage are sharply decreasing, and automated image analyses greatly increase the scale, reproducibility and robustness of biological studies. However, automated image analysis in ecological and evolutionary studies is still in its infancy. There is a clear need for easy to use and accessible tools.
Here, we provide a general purpose method to obtain estimates of population densities, individual body sizes and behavioural metrics from video material of moving organisms. The methods are supplied as a new r‐package trackdem, which provides a flexible, easy to install and use, generally applicable and accurate way to analyse ecological video data. The package can detect and track moving particles, count individuals and estimate individual sizes using background detection, particle identification and particle tracking algorithms. Machine learning is implemented to reduce the influence of noise in lower quality videos or to distinguish a single species in multi‐species systems.
We show that trackdem provides accurate population counts and body size distributions. Using a series of simulations, we show that our estimates are robust against high levels of noise in videos. When applied to live populations of Daphnia magna, our methods obtained accurate and unbiased estimates of population counts, individual sizes and size distributions, as verified by manual counting and measuring. The package trackdem is also directly usable for movement analysis, for instance in behavioural ecology, as illustrated by the tracking of insects, fish, cars and humans.
Within 24 hr, we obtained 192 accurate population counts and body sizes of 22,154 individuals. Such results underscore that automated analysis can improve robustness and reproducibility, and greatly increase the scope of studies in ecology and evolution.</description><subject>automated population counts</subject><subject>Automation</subject><subject>Biological evolution</subject><subject>Body size</subject><subject>Cameras</subject><subject>Computer simulation</subject><subject>Cost analysis</subject><subject>Counting</subject><subject>Data processing</subject><subject>Data storage</subject><subject>Digital cameras</subject><subject>Digital imaging</subject><subject>Ecological monitoring</subject><subject>Ecology</subject><subject>Estimates</subject><subject>Image analysis</subject><subject>Image processing</subject><subject>individual trajectories</subject><subject>Insects</subject><subject>Learning algorithms</subject><subject>Machine learning</subject><subject>movement behaviour</subject><subject>neural net</subject><subject>noise filtering</subject><subject>Noise reduction</subject><subject>particle identification</subject><subject>Particle tracking</subject><subject>Population</subject><subject>Population statistics</subject><subject>Reproducibility</subject><subject>Robustness</subject><subject>size distribution</subject><subject>Video data</subject><issn>2041-210X</issn><issn>2041-210X</issn><fulltext>false</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><recordid>eNqFkMtLxDAQxoMouKx79hrw3N08-ki9Lcv6gBUvCt5CmodkbZuapMr619tuRbw5lxnm-34z8AFwidESD7UiKMUJwehliUlZZCdg9rs5_TOfg0UIezQUZSUi6QzY6IV8U7q5hus-ukZErWAnfLSy1vAo2vYVRgddFYVtYee6vhbRuhZK17cxQNEqGOyXhsqG6G3Vj2KAxrsGflilXYAD5y_AmRF10IufPgfPN9unzV2ye7y936x3iaRZkSWaGlFoZjQmBUlNJbKM0hxpygzJcp2rSkqM0rSiuSoIyytkRC4VopKxgjKK5uBqutt5997rEPne9b4dXnKCCClTxMpycK0ml_QuBK8N77xthD9wjPgYKR9D42No_BjpQOQT8WlrffjPzh-2WzqB30VDeZ4</recordid><startdate>201804</startdate><enddate>201804</enddate><creator>Bruijning, Marjolein</creator><creator>Visser, Marco D.</creator><creator>Hallmann, Caspar A.</creator><creator>Jongejans, Eelke</creator><creator>Golding, Nick</creator><general>John Wiley & Sons, Inc</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7QG</scope><scope>7SN</scope><scope>8FD</scope><scope>C1K</scope><scope>FR3</scope><scope>P64</scope><scope>RC3</scope><orcidid>https://orcid.org/0000-0002-4630-0522</orcidid><orcidid>https://orcid.org/0000-0002-9408-2187</orcidid><orcidid>https://orcid.org/0000-0003-1200-0852</orcidid><orcidid>https://orcid.org/0000-0003-1148-7419</orcidid></search><sort><creationdate>201804</creationdate><title>trackdem: Automated particle tracking to obtain population counts and size distributions from videos in r</title><author>Bruijning, Marjolein ; Visser, Marco D. ; Hallmann, Caspar A. ; Jongejans, Eelke ; Golding, Nick</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>automated population counts</topic><topic>Automation</topic><topic>Biological evolution</topic><topic>Body size</topic><topic>Cameras</topic><topic>Computer simulation</topic><topic>Cost analysis</topic><topic>Counting</topic><topic>Data processing</topic><topic>Data storage</topic><topic>Digital cameras</topic><topic>Digital imaging</topic><topic>Ecological monitoring</topic><topic>Ecology</topic><topic>Estimates</topic><topic>Image analysis</topic><topic>Image processing</topic><topic>individual trajectories</topic><topic>Insects</topic><topic>Learning algorithms</topic><topic>Machine learning</topic><topic>movement behaviour</topic><topic>neural net</topic><topic>noise filtering</topic><topic>Noise reduction</topic><topic>particle identification</topic><topic>Particle tracking</topic><topic>Population</topic><topic>Population statistics</topic><topic>Reproducibility</topic><topic>Robustness</topic><topic>size distribution</topic><topic>Video data</topic><toplevel>peer_reviewed</toplevel><creatorcontrib>Bruijning, Marjolein</creatorcontrib><creatorcontrib>Visser, Marco D.</creatorcontrib><creatorcontrib>Hallmann, Caspar A.</creatorcontrib><creatorcontrib>Jongejans, Eelke</creatorcontrib><creatorcontrib>Golding, Nick</creatorcontrib><collection>CrossRef</collection><collection>Animal Behavior Abstracts</collection><collection>Ecology Abstracts</collection><collection>Technology Research Database</collection><collection>Environmental Sciences and Pollution Management</collection><collection>Engineering Research Database</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>Genetics Abstracts</collection><jtitle>Methods in ecology and evolution</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>no_fulltext</fulltext></delivery><addata><au>Bruijning, Marjolein</au><au>Visser, Marco D.</au><au>Hallmann, Caspar A.</au><au>Jongejans, Eelke</au><au>Golding, Nick</au><au>Golding, Nick</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>trackdem: Automated particle tracking to obtain population counts and size distributions from videos in r</atitle><jtitle>Methods in ecology and evolution</jtitle><date>2018-04</date><risdate>2018</risdate><volume>9</volume><issue>4</issue><spage>965</spage><epage>973</epage><pages>965-973</pages><issn>2041-210X</issn><eissn>2041-210X</eissn><abstract>The possibilities for image analysis in scientific research are substantial: the costs of digital cameras and data storage are sharply decreasing, and automated image analyses greatly increase the scale, reproducibility and robustness of biological studies. However, automated image analysis in ecological and evolutionary studies is still in its infancy. There is a clear need for easy to use and accessible tools.
Here, we provide a general purpose method to obtain estimates of population densities, individual body sizes and behavioural metrics from video material of moving organisms. The methods are supplied as a new r‐package trackdem, which provides a flexible, easy to install and use, generally applicable and accurate way to analyse ecological video data. The package can detect and track moving particles, count individuals and estimate individual sizes using background detection, particle identification and particle tracking algorithms. Machine learning is implemented to reduce the influence of noise in lower quality videos or to distinguish a single species in multi‐species systems.
We show that trackdem provides accurate population counts and body size distributions. Using a series of simulations, we show that our estimates are robust against high levels of noise in videos. When applied to live populations of Daphnia magna, our methods obtained accurate and unbiased estimates of population counts, individual sizes and size distributions, as verified by manual counting and measuring. The package trackdem is also directly usable for movement analysis, for instance in behavioural ecology, as illustrated by the tracking of insects, fish, cars and humans.
Within 24 hr, we obtained 192 accurate population counts and body sizes of 22,154 individuals. Such results underscore that automated analysis can improve robustness and reproducibility, and greatly increase the scope of studies in ecology and evolution.</abstract><cop>London</cop><pub>John Wiley & Sons, Inc</pub><doi>10.1111/2041-210X.12975</doi><tpages>9</tpages><orcidid>https://orcid.org/0000-0002-4630-0522</orcidid><orcidid>https://orcid.org/0000-0002-9408-2187</orcidid><orcidid>https://orcid.org/0000-0003-1200-0852</orcidid><orcidid>https://orcid.org/0000-0003-1148-7419</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | no_fulltext |
| identifier | ISSN: 2041-210X |
| ispartof | Methods in ecology and evolution, 2018-04, Vol.9 (4), p.965-973 |
| issn | 2041-210X 2041-210X |
| language | eng |
| recordid | cdi_proquest_journals_2022940899 |
| source | |
| subjects | automated population counts Automation Biological evolution Body size Cameras Computer simulation Cost analysis Counting Data processing Data storage Digital cameras Digital imaging Ecological monitoring Ecology Estimates Image analysis Image processing individual trajectories Insects Learning algorithms Machine learning movement behaviour neural net noise filtering Noise reduction particle identification Particle tracking Population Population statistics Reproducibility Robustness size distribution Video data |
| title | trackdem: Automated particle tracking to obtain population counts and size distributions from videos in r |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-11-13T07%3A27%3A44IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=trackdem:%20Automated%20particle%20tracking%20to%20obtain%20population%20counts%20and%20size%20distributions%20from%20videos%20in%20r&rft.jtitle=Methods%20in%20ecology%20and%20evolution&rft.au=Bruijning,%20Marjolein&rft.date=2018-04&rft.volume=9&rft.issue=4&rft.spage=965&rft.epage=973&rft.pages=965-973&rft.issn=2041-210X&rft.eissn=2041-210X&rft_id=info:doi/10.1111/2041-210X.12975&rft_dat=%3Cproquest_cross%3E2022940899%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c3575-e3fa7e8fe12724fba553360e38f256e6dbcc1044b36d7286b0fa6cd03c8873830%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2022940899&rft_id=info:pmid/&rfr_iscdi=true |