Tracking droplets in soft granular flows with deep learning techniques

@article{Durve2021TrackingDI,
  title={Tracking droplets in soft granular flows with deep learning techniques},
  author={Mihir Durve and Fabio Bonaccorso and Andrea Montessori and Marco Lauricella and Adriano Tiribocchi and Sauro Succi},
  journal={European Physical Journal plus},
  year={2021},
  volume={136}
}
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… 

DropTrack - automatic droplet tracking using deep learning for microfluidic applications

TLDR
Two deep learning-based cutting-edge algorithms for object detection [you only look once (YOLO)] and object tracking are combined into a single image analysis tool, DropTrack, to track droplets in the microfluidic experiments, which yields the best accuracy in droplet detection and droplet counting in real experimental videos.

Machine learning assisted droplet trajectories extraction in dense emulsions and their analysis

This work analyzes trajectories obtained by YOLO and DeepSORT algorithms of dense emulsion systems simulated by Lattice Boltzmann methods. The results indicate that the individual droplet’s moving

PDC: Pearl Detection with a Counter Based on Deep Learning

TLDR
The robust performance of the proposed method across the pearl dataset indicates that Faster R-CNN ResNet152 with a counter is promising for natural light or artificial light peal detection and accurate counting.

References

SHOWING 1-10 OF 149 REFERENCES

A fast and efficient deep learning procedure for tracking droplet motion in dense microfluidic emulsions

TLDR
A deep learning-based object detection and object tracking algorithm is presented to study droplet motion in dense microfluidic emulsions to correctly predict the droplets’ shape and track their motion at competitive rates as compared to standard clustering algorithms, even in the presence of significant deformations.

Using machine learning to discover shape descriptors for predicting emulsion stability in a microfluidic channel.

TLDR
A machine learning method that learns to discover a low-dimensional code to describe droplet shapes within a concentrated emulsion, and predict whether the drop will become unstable and undergo break-up, and it is observed that 4 out of the 8 dimensions of the code are interpretable, corresponding to drop skewness, elongation, throat size, and surface curvature, respectively.

Efficient Golf Ball Detection and Tracking Based on Convolutional Neural Networks and Kalman Filter

This paper focuses on the problem of online golf ball detection and tracking from image sequences. An efficient real-time approach is proposed by exploiting convolutional neural networks (CNN) based

The vortex-driven dance of droplets within droplets

Understanding the fluid-structure interaction is crucial for an optimal design and manufacturing of soft mesoscale materials. Multi-core emulsions are a class of soft fluids assembled from cluster

The vortex-driven dynamics of droplets within droplets

TLDR
In typical situations expected during microfluidic post-processing, the dynamical distribution of emulsified droplets is dictated by the internal vortices of the host droplet, showing numerical evidence of a surprisingly rich variety of driven non-equilibrium states (NES) within the microchannel.

SSD: Single Shot MultiBox Detector

TLDR
The approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location, which makes SSD easy to train and straightforward to integrate into systems that require a detection component.

Real-time Golf Ball Detection and Tracking Based on Convolutional Neural Networks

TLDR
This paper proposes an efficient and effective solution by integrating object detection and a discrete Kalman model for real-time detection and tracking of a golf ball from video sequences using image patches rather than the entire images for detection.

Simple online and realtime tracking with a deep association metric

TLDR
This paper integrates appearance information to improve the performance of SORT and reduces the number of identity switches, achieving overall competitive performance at high frame rates.

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

TLDR
This work introduces a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals and further merge RPN and Fast R-CNN into a single network by sharing their convolutionAL features.

Object Detection Using Convolutional Neural Networks

TLDR
Two state of the art models are compared for object detection, Single Shot Multi-Box Detector with MobileNetV1 and a Faster Region-based Convolutional Neural Network (Faster-RCNN) with InceptionV2, and result shows that one model is ideal for real-time application because of speed and the other can be used for more accurate object detection.
...