[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122592-en":3,"doc-seo-122592-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122592,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","Machine learning for flow field measurements - a perspective","Advancements in machine-learning techniques are reshaping image processing for flow diagnostics using optical methods. Focusing on Particle Image Velocimetry, the perspective reviews the driving forces behind recent ML developments for flow-field measurements and summarizes key progress in the field. It highlights how supervised, unsupervised, and reinforcement learning approaches are applied to improve preprocessing, velocity-field extraction, and enhancement in space and time, while outlining promising directions for future research and methodological refinement.","This is the version of the article before peer review or editing, as submitted by an author to International Journal of Social Robotics.  \nTitle: Machine learning for flow field measurements: a perspective  \nAuthors: Stefano Discetti-Yingzheng Liu  \nVersion of record available online at [https://doi.org/10.1088/1361-6501/ac9991](https://doi.org/10.1088/1361-6501/ac9991)  \nMachine learning for flow field measurements: a perspective  \nStefano Discetti1 and Yingzheng Liu2  \n1 Aerospace Engineering Research Group, Universidad Carlos III de Madrid, Leganés, Spain  \n2 Gas Turbine Research Institute/School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai, China  \nE-mail: [sdiscett@ing.uc3m.es](sdiscett@ing.uc3m.es), [yzliu@sjtu.edu.cn](yzliu@sjtu.edu.cn)  \nAbstract  \nAdvancements in machine-learning techniques are driving a paradigm shift in image processing. Flow diagnostics with optical techniques is not an exception. Considering the existing and foreseeable disruptive developments in flow-field measurement techniques, we elaborate this perspective, particularly focused to the field of Particle Image Velocimetry. The driving forces for the advancements in machine-learning methods for flow diagnostics in recent years are reviewed, and possible routes for further developments are highlighted.  \nKeywords: machine learning, flow-field measurements, image processing, particle image velocimetry  \n1 Introduction  \nMachine learning (ML) is a subfield of artificial intelligence that aims at using data to perform tasks without human intervention. Examples of such tasks are pattern identification, predictive analytics, data mining, and filtering. In recent years, ML techniques have had a disruptive effect on a wide variety of scientific and engineering fields. The widespread development of ML methods has led to remarkable advancements in several disciplines, disclosing new research pathways whose merit is clearly apparent. ML encompasses a range of techniques that can be classified into three groups:  \n▪ Supervised learning, which generally is aimed at identifying the mapping between a set of input data with (discrete or continuous) output labels. Classification is a classic example of a task performed using supervised learning methods. An artificial neural network (ANN) can be trained using labeled data to identify the class to which an unlabeled input belongs. Supervised learning is also often used for regression tasks, in which the value of an output variable is determined based on statistical correlation with input parameters.  \n▪ Unsupervised learning, which identifies relations between data without any form of supervision. Such techniques include association and clustering algorithms that group unlabeled input data based on similarities. Data compression and dimensionality-reduction techniques also belong to this group. These techniques leverage the concept that the information contained in input data typically lies in a low-dimensional space. Additionally, unsupervised learning techniques include generative models that generate samples after having observed the statistical distribution of training data.  \n▪ Reinforcement learning, in which an agent learns, through its interactions with an environment, strategies to maximize rewards and minimize penalties according to a metric associated with a certain goal to be achieved.  \nAmong others, the field of computer vision has significantly benefited from the advancements in ML in the past decade. ML techniques for tasks such as automatic pattern recognition and tracking, image segmentation, filtering and quality enhancement, and super-resolution are being continuously developed. These innovations have been flowing progressively through the wide scientific community working on flow-field measurement techniques, such as particle image velocimetry (PIV), flow visualization, schlieren imaging, particle image thermometry, and any other technique involving the optical measurement ","cbCain6BvywjlgXq","https://ap.wps.com/l/cbCain6BvywjlgXq","pdf",3317320,1,20,"English","en",105,"# Introduction\n## ML fundamentals\n## ML methods applied to flow-field measurements\n# ML methods for PIV image preprocessing\n## Image quality and contrast\n## Traditional preprocessing techniques\n## Automatic unsupervised preprocessing trends\n# Data processing and conditioning (overview)\n## Direct velocity-field extraction (Section 3.1)\n## Resolution enhancement in space and time (Sections 3.2-3.3)\n## Data augmentation and filtering (Section 3.4)\n# Concluding remarks","[{\"question\":\"What is the article’s main focus regarding machine learning and flow measurements?\",\"answer\":\"It provides a perspective on how machine-learning techniques are driving changes in optical flow diagnostics, with particular emphasis on Particle Image Velocimetry (PIV).\"},{\"question\":\"Which machine-learning categories are discussed in the introduction?\",\"answer\":\"The introduction groups ML into supervised learning, unsupervised learning, and reinforcement learning, and explains representative goals for each category.\"},{\"question\":\"Why is image preprocessing important for PIV, according to the document?\",\"answer\":\"PIV accuracy depends on image quality; preprocessing improves contrast between particles and background and reduces undesired artifacts such as laser reflections that can affect reconstruction, especially in near-surface measurements.\"}]","Machine learning for flow field measurements - a perspective | PDF",1785811636,50,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-for-flow-field-measurements-a-perspective","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-for-flow-field-measurements-a-perspective/122592/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-04",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What is the article’s main focus regarding machine learning and flow measurements?","Question",{"text":76,"@type":77},"It provides a perspective on how machine-learning techniques are driving changes in optical flow diagnostics, with particular emphasis on Particle Image Velocimetry (PIV).","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which machine-learning categories are discussed in the introduction?",{"text":81,"@type":77},"The introduction groups ML into supervised learning, unsupervised learning, and reinforcement learning, and explains representative goals for each category.",{"name":83,"@type":74,"acceptedAnswer":84},"Why is image preprocessing important for PIV, according to the document?",{"text":85,"@type":77},"PIV accuracy depends on image quality; preprocessing improves contrast between particles and background and reduces undesired artifacts such as laser reflections that can affect reconstruction, especially in near-surface measurements.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,115,120,123,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":29,"slug":114},6,"Technology","technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":21,"slug":126},9,"Religion & Spirituality","religion-spirituality",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":21,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":107,"slug":137},19,"General","general"]