Inga Burke, Thajeevan Dhayaparan, Ahmed S. Youssef, Katharina Schmidt, Norbert Kockmann
{"title":"比较两种深度学习方法以确定乳化流动过程中的液滴大小","authors":"Inga Burke, Thajeevan Dhayaparan, Ahmed S. Youssef, Katharina Schmidt, Norbert Kockmann","doi":"10.1007/s41981-024-00330-3","DOIUrl":null,"url":null,"abstract":"<div><p>For reliable supervision in multiphase processes, the droplet size represents a critical quality attribute and needs to be monitored. A promising approach is the use of smart image flow sensors since optical measurement is the most commonly used technique for droplet size distribution determination. For this, two different AI-based object detection methods, Mask RCNN and YOLOv4, are compared regarding their accuracy and their applicability to an emulsification flow process. Iterative optimization steps, including data diversification and adaption of training parameters, enable the models to achieve robust detection performance across varying image qualities and compositions. YOLOv4 shows better detection performances and more accurate results which leads to a wider application window than Mask RCNN in determining droplet sizes in emulsification processes. The final droplet detection model YOLOv4 with Hough Circle (HC) for feature extraction determines reliable droplet sizes across diverse datasets of liquid-liquid flow systems (disperse phase content 1–15 vol.-%, droplet size range 5–150 μm). Evaluating the adjustment of Confidence Scores (CS) ensures statistical representation of even smaller droplets. The droplet detection performance of the final YOLOv4 model is compared with a manual image processing method to validate the model in general as well as its accuracy and reliability. Since YOLOv4 in combination with Hough Circle (HC) shows an accurate and robust detection and size determination, it is applicable for online monitoring and characterization of various liquid-liquid flow processes.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":630,"journal":{"name":"Journal of Flow Chemistry","volume":"14 4","pages":"597 - 613"},"PeriodicalIF":2.0000,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s41981-024-00330-3.pdf","citationCount":"0","resultStr":"{\"title\":\"Two deep learning methods in comparison to characterize droplet sizes in emulsification flow processes\",\"authors\":\"Inga Burke, Thajeevan Dhayaparan, Ahmed S. Youssef, Katharina Schmidt, Norbert Kockmann\",\"doi\":\"10.1007/s41981-024-00330-3\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>For reliable supervision in multiphase processes, the droplet size represents a critical quality attribute and needs to be monitored. A promising approach is the use of smart image flow sensors since optical measurement is the most commonly used technique for droplet size distribution determination. For this, two different AI-based object detection methods, Mask RCNN and YOLOv4, are compared regarding their accuracy and their applicability to an emulsification flow process. Iterative optimization steps, including data diversification and adaption of training parameters, enable the models to achieve robust detection performance across varying image qualities and compositions. YOLOv4 shows better detection performances and more accurate results which leads to a wider application window than Mask RCNN in determining droplet sizes in emulsification processes. The final droplet detection model YOLOv4 with Hough Circle (HC) for feature extraction determines reliable droplet sizes across diverse datasets of liquid-liquid flow systems (disperse phase content 1–15 vol.-%, droplet size range 5–150 μm). Evaluating the adjustment of Confidence Scores (CS) ensures statistical representation of even smaller droplets. The droplet detection performance of the final YOLOv4 model is compared with a manual image processing method to validate the model in general as well as its accuracy and reliability. Since YOLOv4 in combination with Hough Circle (HC) shows an accurate and robust detection and size determination, it is applicable for online monitoring and characterization of various liquid-liquid flow processes.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>\",\"PeriodicalId\":630,\"journal\":{\"name\":\"Journal of Flow Chemistry\",\"volume\":\"14 4\",\"pages\":\"597 - 613\"},\"PeriodicalIF\":2.0000,\"publicationDate\":\"2024-07-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://link.springer.com/content/pdf/10.1007/s41981-024-00330-3.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Flow Chemistry\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s41981-024-00330-3\",\"RegionNum\":4,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"CHEMISTRY, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Flow Chemistry","FirstCategoryId":"92","ListUrlMain":"https://link.springer.com/article/10.1007/s41981-024-00330-3","RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
Two deep learning methods in comparison to characterize droplet sizes in emulsification flow processes
For reliable supervision in multiphase processes, the droplet size represents a critical quality attribute and needs to be monitored. A promising approach is the use of smart image flow sensors since optical measurement is the most commonly used technique for droplet size distribution determination. For this, two different AI-based object detection methods, Mask RCNN and YOLOv4, are compared regarding their accuracy and their applicability to an emulsification flow process. Iterative optimization steps, including data diversification and adaption of training parameters, enable the models to achieve robust detection performance across varying image qualities and compositions. YOLOv4 shows better detection performances and more accurate results which leads to a wider application window than Mask RCNN in determining droplet sizes in emulsification processes. The final droplet detection model YOLOv4 with Hough Circle (HC) for feature extraction determines reliable droplet sizes across diverse datasets of liquid-liquid flow systems (disperse phase content 1–15 vol.-%, droplet size range 5–150 μm). Evaluating the adjustment of Confidence Scores (CS) ensures statistical representation of even smaller droplets. The droplet detection performance of the final YOLOv4 model is compared with a manual image processing method to validate the model in general as well as its accuracy and reliability. Since YOLOv4 in combination with Hough Circle (HC) shows an accurate and robust detection and size determination, it is applicable for online monitoring and characterization of various liquid-liquid flow processes.
期刊介绍:
The main focus of the journal is flow chemistry in inorganic, organic, analytical and process chemistry in the academic research as well as in applied research and development in the pharmaceutical, agrochemical, fine-chemical, petro- chemical, fragrance industry.