• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

运用深度学习技术追踪软颗粒流中的液滴。

Tracking droplets in soft granular flows with deep learning techniques.

作者信息

Durve Mihir, Bonaccorso Fabio, Montessori Andrea, Lauricella Marco, Tiribocchi Adriano, Succi Sauro

机构信息

Center for Life Nano- & Neuro-Science, Fondazione Istituto Italiano di Tecnologia (IIT), 00161 Rome, Italy.

Istituto per le Applicazioni del Calcolo CNR, via dei Taurini 19, Rome, Italy.

出版信息

Eur Phys J Plus. 2021;136(8):864. doi: 10.1140/epjp/s13360-021-01849-3. Epub 2021 Aug 21.

DOI:10.1140/epjp/s13360-021-01849-3
PMID:34458055
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC8380117/
Abstract

The state-of-the-art deep learning-based object recognition YOLO algorithm and object tracking DeepSORT algorithm are combined to analyze digital images from fluid dynamic simulations of multi-core emulsions and soft flowing crystals and to track moving droplets within these complex flows. The YOLO network was trained to recognize the droplets with synthetically prepared data, thereby bypassing the labor-intensive data acquisition process. In both applications, the trained YOLO + DeepSORT procedure performs with high accuracy on the real data from the fluid simulations, with low error levels in the inferred trajectories of the droplets and independently computed ground truth. Moreover, using commonly used desktop GPUs, the developed application is capable of analyzing data at speeds that exceed the typical image acquisition rates of digital cameras (30 fps), opening the interesting prospect of realizing a low-cost and practical tool to study systems with many moving objects, mostly but not exclusively, biological ones. Besides its practical applications, the procedure presented here marks the first step towards the automatic extraction of effective equations of motion of many-body soft flowing systems.

摘要

将基于深度学习的先进目标识别YOLO算法与目标跟踪DeepSORT算法相结合，用于分析多核乳液和软流动晶体的流体动力学模拟中的数字图像，并跟踪这些复杂流动中的移动液滴。YOLO网络使用合成制备的数据进行训练，以识别液滴，从而绕过了劳动密集型的数据采集过程。在这两个应用中，经过训练的YOLO + DeepSORT程序对来自流体模拟的真实数据具有高精度，液滴推断轨迹和独立计算的地面真值的误差水平较低。此外，使用常用的桌面GPU，所开发的应用程序能够以超过数码相机典型图像采集速率（30帧/秒）的速度分析数据，为实现一种低成本且实用的工具来研究具有许多移动物体（主要但不限于生物物体）的系统开辟了有趣的前景。除了其实际应用外，这里介绍的程序标志着朝着自动提取多体软流动系统有效运动方程迈出的第一步。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/f773056e75ef/13360_2021_1849_Fig15_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/fc6e75e7b6c9/13360_2021_1849_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/7283ff0a5733/13360_2021_1849_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/737c77641183/13360_2021_1849_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/a3aa2c80a29e/13360_2021_1849_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/4ac5d3ad2931/13360_2021_1849_Fig5_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/5386cebc428c/13360_2021_1849_Fig6_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/da3b0a0ca8f7/13360_2021_1849_Fig7_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/57ae770e9ec6/13360_2021_1849_Fig8_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/63ce8e7c2102/13360_2021_1849_Fig9_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/bdb887701617/13360_2021_1849_Fig10_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/eea309f6af34/13360_2021_1849_Fig11_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/8248bd4ad382/13360_2021_1849_Fig12_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/90f2a4f06454/13360_2021_1849_Fig13_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/08a13ba1b107/13360_2021_1849_Fig14_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/f773056e75ef/13360_2021_1849_Fig15_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/fc6e75e7b6c9/13360_2021_1849_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/7283ff0a5733/13360_2021_1849_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/737c77641183/13360_2021_1849_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/a3aa2c80a29e/13360_2021_1849_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/4ac5d3ad2931/13360_2021_1849_Fig5_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/5386cebc428c/13360_2021_1849_Fig6_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/da3b0a0ca8f7/13360_2021_1849_Fig7_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/57ae770e9ec6/13360_2021_1849_Fig8_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/63ce8e7c2102/13360_2021_1849_Fig9_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/bdb887701617/13360_2021_1849_Fig10_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/eea309f6af34/13360_2021_1849_Fig11_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/8248bd4ad382/13360_2021_1849_Fig12_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/90f2a4f06454/13360_2021_1849_Fig13_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/08a13ba1b107/13360_2021_1849_Fig14_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/83e1/8550127/f773056e75ef/13360_2021_1849_Fig15_HTML.jpg

相似文献

1
Tracking droplets in soft granular flows with deep learning techniques.运用深度学习技术追踪软颗粒流中的液滴。
Eur Phys J Plus. 2021;136(8):864. doi: 10.1140/epjp/s13360-021-01849-3. Epub 2021 Aug 21.
2
A fast and efficient deep learning procedure for tracking droplet motion in dense microfluidic emulsions.一种快速高效的深度学习程序，用于跟踪密集微流控乳液中的液滴运动。
Philos Trans A Math Phys Eng Sci. 2021 Oct 18;379(2208):20200400. doi: 10.1098/rsta.2020.0400. Epub 2021 Aug 30.
3
Benchmarking YOLOv5 and YOLOv7 models with DeepSORT for droplet tracking applications.基准测试 DeepSORT 与 YOLOv5 和 YOLOv7 模型在液滴跟踪应用中的性能。
Eur Phys J E Soft Matter. 2023 May 8;46(5):32. doi: 10.1140/epje/s10189-023-00290-x.
4
A Robust Real-Time Detecting and Tracking Framework for Multiple Kinds of Unmarked Object.一种用于多种无标记物体的稳健实时检测与跟踪框架。
Sensors (Basel). 2019 Dec 18;20(1):2. doi: 10.3390/s20010002.
5
Real-Time Pattern-Recognition of GPR Images with YOLO v3 Implemented by Tensorflow.基于 Tensorflow 实现的 YOLO v3 的探地雷达图像实时模式识别。
Sensors (Basel). 2020 Nov 12;20(22):6476. doi: 10.3390/s20226476.
6
Self-contained deep learning-based boosting of 4D cone-beam CT reconstruction.基于深度学习的独立式4D锥形束CT重建增强技术
Med Phys. 2020 Nov;47(11):5619-5631. doi: 10.1002/mp.14441. Epub 2020 Oct 15.
7
Macromolecular crowding: chemistry and physics meet biology (Ascona, Switzerland, 10-14 June 2012).大分子拥挤现象：化学与物理邂逅生物学（瑞士阿斯科纳，2012年6月10日至14日）
Phys Biol. 2013 Aug;10(4):040301. doi: 10.1088/1478-3975/10/4/040301. Epub 2013 Aug 2.
8
Parking Time Violation Tracking Using YOLOv8 and Tracking Algorithms.使用 YOLOv8 和跟踪算法进行停车时间违规跟踪。
Sensors (Basel). 2023 Jun 23;23(13):5843. doi: 10.3390/s23135843.
9
Mobile Eye-Tracking Data Analysis Using Object Detection via YOLO v4.基于 YOLO v4 的目标检测的移动眼动追踪数据分析。
Sensors (Basel). 2021 Nov 18;21(22):7668. doi: 10.3390/s21227668.
10
Attention-aware fully convolutional neural network with convolutional long short-term memory network for ultrasound-based motion tracking.基于注意力感知的全卷积神经网络与卷积长短期记忆网络的超声运动跟踪
Med Phys. 2019 May;46(5):2275-2285. doi: 10.1002/mp.13510. Epub 2019 Apr 22.

引用本文的文献

1
PDC: Pearl Detection with a Counter Based on Deep Learning.PDC：基于深度学习的珍珠检测。
Sensors (Basel). 2022 Sep 16;22(18):7026. doi: 10.3390/s22187026.

本文引用的文献

1
A fast and efficient deep learning procedure for tracking droplet motion in dense microfluidic emulsions.一种快速高效的深度学习程序，用于跟踪密集微流控乳液中的液滴运动。
Philos Trans A Math Phys Eng Sci. 2021 Oct 18;379(2208):20200400. doi: 10.1098/rsta.2020.0400. Epub 2021 Aug 30.
2
Can deep learning beat numerical weather prediction?深度学习能打败数值天气预报吗？
Philos Trans A Math Phys Eng Sci. 2021 Apr 5;379(2194):20200097. doi: 10.1098/rsta.2020.0097. Epub 2021 Feb 15.
3
The vortex-driven dynamics of droplets within droplets.
液滴内液滴的涡旋驱动动力学。
Nat Commun. 2021 Jan 4;12(1):82. doi: 10.1038/s41467-020-20364-0.
4
Improved Handwritten Digit Recognition Using Convolutional Neural Networks (CNN).基于卷积神经网络（CNN）的改进手写数字识别。
Sensors (Basel). 2020 Jun 12;20(12):3344. doi: 10.3390/s20123344.
5
Highly ordered and tunable polyHIPEs by using microfluidics.通过微流控技术制备高度有序且可调节的聚（高内相乳液）
J Mater Chem B. 2014 Apr 28;2(16):2290-2300. doi: 10.1039/c3tb21227k. Epub 2014 Mar 17.
6
Improved protein structure prediction using potentials from deep learning.利用深度学习势进行蛋白质结构预测的改进。
Nature. 2020 Jan;577(7792):706-710. doi: 10.1038/s41586-019-1923-7. Epub 2020 Jan 15.
7
Comprehensive comparison of collision models in the lattice Boltzmann framework: Theoretical investigations.晶格 Boltzmann 框架中的碰撞模型综合比较：理论研究。
Phys Rev E. 2019 Sep;100(3-1):033305. doi: 10.1103/PhysRevE.100.033305.
8
Lattice Boltzmann methods and active fluids.格子玻尔兹曼方法与活性流体
Eur Phys J E Soft Matter. 2019 Jun 28;42(6):81. doi: 10.1140/epje/i2019-11843-6.
9
A Computer Vision-Based Roadside Occupation Surveillance System for Intelligent Transport in Smart Cities.一种用于智慧城市智能交通的基于计算机视觉的路边职业监测系统。
Sensors (Basel). 2019 Apr 15;19(8):1796. doi: 10.3390/s19081796.
10
Big data: the end of the scientific method?大数据：科学方法的终结？
Philos Trans A Math Phys Eng Sci. 2019 Apr 8;377(2142):20180145. doi: 10.1098/rsta.2018.0145.