Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing
<p dir="ltr">Food computing has emerged as a promising research field, employing artificial intelligence, deep learning, and data science methodologies to enhance various stages of food production pipelines. To this end, the food computing community has compiled a variety of data set...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , , |
| منشور في: |
2024
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1864513541533335552 |
|---|---|
| author | Nauman Ullah Gilal (17302714) |
| author2 | Marwa Qaraqe (10135172) Jens Schneider (16885948) Marco Agus (8032898) |
| author2_role | author author author |
| author_facet | Nauman Ullah Gilal (17302714) Marwa Qaraqe (10135172) Jens Schneider (16885948) Marco Agus (8032898) |
| author_role | author |
| dc.creator.none.fl_str_mv | Nauman Ullah Gilal (17302714) Marwa Qaraqe (10135172) Jens Schneider (16885948) Marco Agus (8032898) |
| dc.date.none.fl_str_mv | 2024-07-09T06:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1007/s00371-024-03560-7 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Autocleandeepfood_auto-cleaning_and_data_balancing_transfer_learning_for_regional_gastronomy_food_computing/29899604 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Agricultural, veterinary and food sciences Food sciences Information and computing sciences Artificial intelligence Human-centred computing Machine learning Food computing Web-scrapping Traditional cuisines MENA food data set Noisy labels Transfer learning Auto-cleaning Data imbalance |
| dc.title.none.fl_str_mv | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Food computing has emerged as a promising research field, employing artificial intelligence, deep learning, and data science methodologies to enhance various stages of food production pipelines. To this end, the food computing community has compiled a variety of data sets and developed various deep-learning architectures to perform automatic classification. However, automated food classification presents a significant challenge, particularly when it comes to local and regional cuisines, which are often underrepresented in available public-domain data sets. Nevertheless, obtaining high-quality, well-labeled, and well-balanced real-world labeled images is challenging since manual data curation requires significant human effort and is time-consuming. In contrast, the web has a potentially unlimited source of food data but tapping into this resource has a good chance of corrupted and wrongly labeled images. In addition, the uneven distribution among food categories may lead to data imbalance problems. All these issues make it challenging to create clean data sets for food from web data. To address this issue, we present <i>AutoCleanDeepFood</i>, a novel end-to-end food computing framework for regional gastronomy that contains the following components: (i) a fully automated pre-processing pipeline for custom data sets creation related to specific regional gastronomy, (ii) a transfer learning-based training paradigm to filter out noisy labels through loss ranking, incorporating a Russian Roulette probabilistic approach to mitigate data imbalance problems, and (iii) a method for deploying the resulting model on smartphones for real-time inferences. We assess the performance of our framework on a real-world noisy public domain data set, ETH Food-101, and two novel web-collected datasets, MENA-150 and Pizza-Styles. We demonstrate the filtering capabilities of our proposed method through embedding visualization of the feature space using the t-SNE dimension reduction scheme. Our filtering scheme is efficient and effectively improves accuracy in all cases, boosting performance by 0.96, 0.71, and 1.29% on MENA-150, ETH Food-101, and Pizza-Styles, respectively.</p><h2>Other Information</h2><p dir="ltr">Published in: The Visual Computer<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1007/s00371-024-03560-7" target="_blank">https://dx.doi.org/10.1007/s00371-024-03560-7</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_2b8f3b337720a86c811bdb3e9b5235f8 |
| identifier_str_mv | 10.1007/s00371-024-03560-7 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/29899604 |
| publishDate | 2024 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computingNauman Ullah Gilal (17302714)Marwa Qaraqe (10135172)Jens Schneider (16885948)Marco Agus (8032898)Agricultural, veterinary and food sciencesFood sciencesInformation and computing sciencesArtificial intelligenceHuman-centred computingMachine learningFood computingWeb-scrappingTraditional cuisinesMENA food data setNoisy labelsTransfer learningAuto-cleaningData imbalance<p dir="ltr">Food computing has emerged as a promising research field, employing artificial intelligence, deep learning, and data science methodologies to enhance various stages of food production pipelines. To this end, the food computing community has compiled a variety of data sets and developed various deep-learning architectures to perform automatic classification. However, automated food classification presents a significant challenge, particularly when it comes to local and regional cuisines, which are often underrepresented in available public-domain data sets. Nevertheless, obtaining high-quality, well-labeled, and well-balanced real-world labeled images is challenging since manual data curation requires significant human effort and is time-consuming. In contrast, the web has a potentially unlimited source of food data but tapping into this resource has a good chance of corrupted and wrongly labeled images. In addition, the uneven distribution among food categories may lead to data imbalance problems. All these issues make it challenging to create clean data sets for food from web data. To address this issue, we present <i>AutoCleanDeepFood</i>, a novel end-to-end food computing framework for regional gastronomy that contains the following components: (i) a fully automated pre-processing pipeline for custom data sets creation related to specific regional gastronomy, (ii) a transfer learning-based training paradigm to filter out noisy labels through loss ranking, incorporating a Russian Roulette probabilistic approach to mitigate data imbalance problems, and (iii) a method for deploying the resulting model on smartphones for real-time inferences. We assess the performance of our framework on a real-world noisy public domain data set, ETH Food-101, and two novel web-collected datasets, MENA-150 and Pizza-Styles. We demonstrate the filtering capabilities of our proposed method through embedding visualization of the feature space using the t-SNE dimension reduction scheme. Our filtering scheme is efficient and effectively improves accuracy in all cases, boosting performance by 0.96, 0.71, and 1.29% on MENA-150, ETH Food-101, and Pizza-Styles, respectively.</p><h2>Other Information</h2><p dir="ltr">Published in: The Visual Computer<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1007/s00371-024-03560-7" target="_blank">https://dx.doi.org/10.1007/s00371-024-03560-7</a></p>2024-07-09T06:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1007/s00371-024-03560-7https://figshare.com/articles/journal_contribution/Autocleandeepfood_auto-cleaning_and_data_balancing_transfer_learning_for_regional_gastronomy_food_computing/29899604CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/298996042024-07-09T06:00:00Z |
| spellingShingle | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing Nauman Ullah Gilal (17302714) Agricultural, veterinary and food sciences Food sciences Information and computing sciences Artificial intelligence Human-centred computing Machine learning Food computing Web-scrapping Traditional cuisines MENA food data set Noisy labels Transfer learning Auto-cleaning Data imbalance |
| status_str | publishedVersion |
| title | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| title_full | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| title_fullStr | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| title_full_unstemmed | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| title_short | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| title_sort | Autocleandeepfood: auto-cleaning and data balancing transfer learning for regional gastronomy food computing |
| topic | Agricultural, veterinary and food sciences Food sciences Information and computing sciences Artificial intelligence Human-centred computing Machine learning Food computing Web-scrapping Traditional cuisines MENA food data set Noisy labels Transfer learning Auto-cleaning Data imbalance |