{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,13]],"date-time":"2025-10-13T09:20:03Z","timestamp":1760347203538,"version":"3.41.0"},"reference-count":47,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2025,5,22]],"date-time":"2025-05-22T00:00:00Z","timestamp":1747872000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,5,22]],"date-time":"2025-05-22T00:00:00Z","timestamp":1747872000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"M\u00e9trologie fran\u00e7aise"},{"name":"Laboratoire national de m\u00e9trologie et d\u2019essais"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,6,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Instance segmentation model performances is commonly evaluated through challenges using toy datasets such as CIFAR and ImageNet. Although, it is necessary to help improving state-of-the-art models, it remains far from critical industrial application where accurate measurement is required to ensure optimal process control, quality assurance and safety. This paper compares the performance of ready-to-deploy state-of-the-art instance segmentation models, specifically region and query-based models, across COCO and task-specific designed metrics in the context of the estimation of aggregated TiO<jats:sub>2<\/jats:sub> particle size distribution from scanning electron microscopy (SEM) measurements. The numerical comparison is conducted on custom datasets, addressing well-known challenges of SEM microscopists: high resolution images showing a massive number of overlapping instances, images measured with diverse imaging parameters (operator control) or the presence of a mixture of aggregated particles. Our findings highlight the strengths and limitations of each model category, providing insights into their suitability for high-stakes industrial deployment. Finally, we propose a customized inference procedure that leverages the performance of transformers on images that contain a small number of instances within dense and clustered scenes.<\/jats:p>","DOI":"10.1088\/2632-2153\/add239","type":"journal-article","created":{"date-parts":[[2025,4,29]],"date-time":"2025-04-29T22:55:50Z","timestamp":1745967350000},"page":"020502","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Region versus query based instance segmentation models: application to the estimation of aggregated TiO<sub>2<\/sub> particles size distribution measured by SEM"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6658-1798","authenticated-orcid":true,"given":"Paul","family":"Monchot","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5834-252X","authenticated-orcid":true,"given":"Lo\u00efc","family":"Coquelin","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3628-8117","authenticated-orcid":true,"given":"Nicolas","family":"Fischer","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,5,22]]},"reference":[{"key":"mlstadd239bib1","doi-asserted-by":"publisher","first-page":"361","DOI":"10.7736\/JKSPE.019.144","article-title":"Quantitative metallographic analysis of gcr15 microstructure using mask r-cnn","volume":"37","author":"Agbozo","year":"2020","journal-title":"J. Korean Soc. Precis. Eng."},{"article-title":"Beit: Bert pre-training of image transformers","year":"2021","author":"Bao","key":"mlstadd239bib2"},{"key":"mlstadd239bib3","doi-asserted-by":"publisher","first-page":"1226","DOI":"10.1038\/s41592-019-0582-9","article-title":"ilastik: interactive machine learning for (bio)image analysis","volume":"16","author":"Berg","year":"2019","journal-title":"Nat. Methods"},{"key":"mlstadd239bib4","first-page":"pp 6154","article-title":"Cascade r-cnn: delving into high quality object detection","author":"Cai","year":"2018"},{"key":"mlstadd239bib5","first-page":"pp 213","article-title":"End-to-end object detection with transformers","author":"Carion","year":"2020"},{"key":"mlstadd239bib6","first-page":"pp 9650","article-title":"Emerging properties in self-supervised vision transformers","author":"Caron","year":"2021"},{"key":"mlstadd239bib7","first-page":"pp 1824","article-title":"Assessing yolact++ for real time and robust instance segmentation of medical instruments in endoscopic procedures","author":"Cer\u00f3n","year":"2021"},{"key":"mlstadd239bib8","first-page":"pp 4974","article-title":"Hybrid task cascade for instance segmentation","author":"Chen","year":"2019"},{"key":"mlstadd239bib9","first-page":"pp 1290","article-title":"Masked-attention mask transformer for universal image segmentation","author":"Cheng","year":"2022"},{"key":"mlstadd239bib10","first-page":"pp 17864","article-title":"Per-pixel classification is not all you need for semantic segmentation","volume":"vol 34","author":"Cheng","year":"2021"},{"article-title":"An image is worth 16x16 words: transformers for image recognition at scale","year":"2020","author":"Dosovitskiy","key":"mlstadd239bib11"},{"key":"mlstadd239bib12","first-page":"pp 6910","article-title":"Instances as queries","author":"Fang","year":"2021"},{"key":"mlstadd239bib13","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1023\/B:VISI.0000022288.19776.77","article-title":"Efficient graph-based image segmentation","volume":"59","author":"Felzenszwalb","year":"2004","journal-title":"Int. J. Comput. Vis."},{"key":"mlstadd239bib14","first-page":"pp 405","article-title":"Towards robust part-aware instance segmentation for industrial bin picking","author":"Feng","year":"2022"},{"key":"mlstadd239bib15","first-page":"pp 1440","article-title":"Fast r-cnn","author":"Girshick","year":"2015"},{"key":"mlstadd239bib16","first-page":"pp 580","article-title":"Rich feature hierarchies for accurate object detection and semantic segmentation","author":"Girshick","year":"2014"},{"key":"mlstadd239bib17","doi-asserted-by":"publisher","DOI":"10.1016\/j.imavis.2022.104401","article-title":"A review on 2d instance segmentation based on deep neural networks","volume":"120","author":"Gu","year":"2022","journal-title":"Image Vis. Comput."},{"key":"mlstadd239bib18","first-page":"pp 2961","article-title":"Mask r-cnn","author":"He","year":"2017"},{"key":"mlstadd239bib19","first-page":"pp 770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"article-title":"Istr: end-to-end instance segmentation with transformers","year":"2021","author":"Hu","key":"mlstadd239bib20"},{"key":"mlstadd239bib21","first-page":"pp 4700","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"key":"mlstadd239bib22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3505244","article-title":"Transformers in vision: a survey","volume":"54","author":"Khan","year":"2022","journal-title":"ACM Comput. Surveys"},{"key":"mlstadd239bib23","first-page":"pp 13619","article-title":"Dn-detr: accelerate detr training by introducing query denoising","author":"Li","year":"2022"},{"key":"mlstadd239bib24","first-page":"pp 3041","article-title":"Mask dino: Towards a unified transformer-based framework for object detection and segmentation","author":"Li","year":"2023"},{"key":"mlstadd239bib25","doi-asserted-by":"publisher","first-page":"6893","DOI":"10.1109\/TIP.2022.3216771","article-title":"Cbnet: a composite backbone network architecture for object detection","volume":"31","author":"Liang","year":"2022","journal-title":"IEEE Trans. Image Process."},{"key":"mlstadd239bib26","first-page":"pp 740","article-title":"Microsoft coco: common objects in context","author":"Lin","year":"2014"},{"article-title":"Dab-detr: dynamic anchor boxes are better queries for detr","year":"2022","author":"Liu","key":"mlstadd239bib27"},{"key":"mlstadd239bib28","first-page":"pp 10012","article-title":"Swin transformer: hierarchical vision transformer using shifted windows","author":"Liu","year":"2021"},{"key":"mlstadd239bib29","doi-asserted-by":"publisher","first-page":"968","DOI":"10.3390\/nano11040968","article-title":"Deep learning based instance segmentation of titanium dioxide particles in the form of agglomerates in scanning electron microscopy","volume":"11","author":"Monchot","year":"2021","journal-title":"Nanomaterials"},{"key":"mlstadd239bib30","doi-asserted-by":"publisher","DOI":"10.1155\/2015\/790508","article-title":"Size measurement of nanoparticle assembly using multilevel segmented tem images","volume":"2015","author":"Muneesawang","year":"2015","journal-title":"J. Nanomater."},{"key":"mlstadd239bib31","first-page":"pp 850","article-title":"Efficient non-maximum suppression","volume":"vol 3","author":"Neubeck","year":"2006"},{"year":"1997","author":"Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA","key":"mlstadd239bib32"},{"key":"mlstadd239bib33","first-page":"pp 779","article-title":"You only look once: unified, real-time object detection","author":"Redmon","year":"2016"},{"key":"mlstadd239bib34","article-title":"Faster r-cnn: towards real-time object detection with region proposal networks","volume":"vol 28","author":"Ren","year":"2015"},{"key":"mlstadd239bib35","first-page":"pp 234","article-title":"U-net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"article-title":"Mean average precision (map) explained: Everything you need to know","year":"2022","author":"Shah","key":"mlstadd239bib36"},{"key":"mlstadd239bib37","doi-asserted-by":"publisher","first-page":"499","DOI":"10.1007\/s42979-022-01407-3","article-title":"A survey on object instance segmentation","volume":"3","author":"Sharma","year":"2022","journal-title":"SN Comput. Sci."},{"article-title":"Very deep convolutional networks for large-scale image recognition","year":"2014","author":"Simonyan","key":"mlstadd239bib38"},{"key":"mlstadd239bib39","first-page":"1","article-title":"A method of establishing groups of equal amplitude in plant sociology based on similarity of species content and its application to analyses of the vegetation on danish commons","volume":"5","author":"Sorensen","year":"1948","journal-title":"Biol. Skar."},{"key":"mlstadd239bib40","first-page":"pp 14454","article-title":"Sparse r-cnn: end-to-end object detection with learnable proposals","author":"Sun","year":"2021"},{"key":"mlstadd239bib41","first-page":"pp 1","article-title":"Going deeper with convolutions","author":"Szegedy","year":"2015"},{"key":"mlstadd239bib42","article-title":"Attention is all you need","volume":"vol 30","author":"Vaswani","year":"2017"},{"key":"mlstadd239bib43","first-page":"pp 232","article-title":"Nanoparticle size measurement method based on improved watershed segmentation","author":"Wu","year":"2018"},{"key":"mlstadd239bib44","first-page":"pp 1492","article-title":"Aggregated residual transformations for deep neural networks","author":"Xie","year":"2017"},{"key":"mlstadd239bib45","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2021.3124835","article-title":"Merged-sampling mask r-cnn with random proposal expansion for particle measurement of sem images of molecular sieve catalysts","volume":"70","author":"Xin","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"article-title":"Dino: detr with improved denoising anchor boxes for end-to-end object detection","year":"2022","author":"Zhang","key":"mlstadd239bib46"},{"key":"mlstadd239bib47","first-page":"pp 669","article-title":"Instance-level segmentation for autonomous driving with deep densely connected mrfs","author":"Zhang","year":"2016"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,22]],"date-time":"2025-05-22T07:35:16Z","timestamp":1747899316000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/add239"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,22]]},"references-count":47,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,5,22]]},"published-print":{"date-parts":[[2025,6,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/add239","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2025,5,22]]},"assertion":[{"value":"Region versus query based instance segmentation models: application to the estimation of aggregated TiO2 particles size distribution measured by SEM","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 Laboratoire national de m\u00e9trologie et d\u2019essais. Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2024-10-08","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-04-29","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-05-22","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}