Particle detection and size recognition based on defocused particle images - a comparison of a deterministic algorithm and a deep neural network

Oct 29, 2026

Upload by Fishy

Research focuses on particle size recognition and volumetric position determination for microfluidic separation and mixing. The study extends astigmatism particle tracking velocimetry with both a deterministic algorithm and a deep neural network to classify particles from multimodal size distributions, achieving up to 99.9% precision for bimodal mixtures without adapting the measurement setup. Detection performance reaches recall up to 99.0%. Semi-synthetic images with consistent ground truth validate correct detection and classification, and the approach generalizes to up to four particle species with high precision and recall, enabling quantitative evaluation of microfluidic processes.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

Introduction

Experiments in Fluids (2023) 64:21 [https://doi.org/10.1007/s00348-023-03574-2](https://doi.org/10.1007/s00348-023-03574-2) Particle detection and size recognition based on defocused particle images: a comparison of a deterministic algorithm and a deep neural network Sebastian Sachs1 · Manuel Ratz1 · Patrick Mäder2 · Jörg König1 · Christian Cierpka1 Received: 15 September 2022 / Revised: 14 October 2022 / Accepted: 6 January 2023 © The Author(s) 2023 Abstract The systematic manipulation of components of multimodal particle solutions is a key for the design of modern industrial products and pharmaceuticals with highly customized properties. In order to optimize innovative particle separation deviceson microfluidic scales, a particle size recognition with simultaneous volumetric position determination is essential. In the present study, the astigmatism particle tracking velocimetry is extended by a deterministic algorithm and a deep neural network (DNN) to include size classification of particles of multimodal size distribution. Without any adaptation of the existing measurement setup, a reliable classification of bimodal particle solutions in the size range of 􀀞 􀀝 μ􀀜– 􀀞􀀝􀀜 μ􀀛 is demonstrated with a precision of up to 99.9 %. Concurrently, the high detection rate of the particles, suspended in a laminar fluid flow, is quantified by a recall of 99.0 % . By extracting particle images from the experimentally acquired images and placing them on a synthetic background, semi-synthetic images with consistent ground truth are generated. These contain labeled overlapping particle images that are correctly detected and classified by the DNN. The study is complemented by employing the presented algorithms for simultaneous size recognition of up to four particle species with a particle diameter in between 􀀞 􀀝 μ􀀜 and 􀀞􀀝􀀜 μ􀀛 . With the very high precision of up to 99.3 % at a recall of 94.8 %, the applicability to classify multimodal particle mixtures even in dense solutions is confirmed. The present contribution thus paves the way for quantitative evaluation of microfluidic separation and mixing processes. 1 Introduction The mixing or separation of microparticles with various sizes is of crucial importance for biological or chemical analysis as well as for customizing pharmaceutical products, paints and coatings (Sajeesh and Sen 2014 ; Hesselet al. 2005) . Innovative processes based on microfluidic lab-on-a-chip devices have achieved major advances for the systematic manipulation of the composition of particle solutions (Ahmad et al. 2019 ; Gossett et al. 2010 ; Miyagawa and Okada 2021; Zhang et al. 2020 ; Zoupanou et al. 2021) . In order to evaluate the performance and further optimize these processes, a particle size recognition is essential. * Sebastian Sachs [sebastian.sachs@tu-ilmenau.de](sebastian.sachs@tu-ilmenau.de) 1 Institute of Thermodynamics and Fluid Mechanics, Technische Universität Ilmenau, 98693 Ilmenau, Germany 2 Data-Intensive Systems and Visualization Group (dAI.SY), Technische Universität Ilmenau, 98693 Ilmenau, Germany Commercially available particle size analysis devices are typically designed for the separate examination of samples obtained from outlets (Shekunov et al. 2007 ; Gross-Rother et al. 2020) . Common devices mostly utilize particle-light interactions (Stetefeld et al. 2016 ; Dannhauser et al. 2014 ; Xu 2015), resistive pulse sensing (Zhang et al. 2019 ; Weatherall and Willmott 2015 ; Grabarek et al. 2019 ; Caputo et al. 2019), suspended microchannel resonators (GrossRother et al. 2020 ; Stockslager et al. 2019 ; de Pastina and Villanueva 2020), or microscopic imaging (Sharma et al. 2010 ; Rice et al. 2013) . Among the available techniques, the microfluidic flow cytometry emerged as a widely used tool for the analysis of suspended particle mixtures (Mach et al. 2011 ; Picot et al. 2012 ; Gong et al. 2018 ; McKinnon 2018) . By means of hydrodynamic focusing, the particles are initially g

FAQs

What problem does the study address in microfluidic particle analysis?

It addresses the need for particle size recognition combined with simultaneous volumetric position determination to optimize microfluidic separation devices and processes involving multimodal particle solutions.

How does the paper extend astigmatism particle tracking velocimetry?

It extends astigmatism particle tracking velocimetry by adding both a deterministic algorithm and a deep neural network to perform size classification of particles from multimodal size distributions.

What validation method is used to assess detection and classification?

Particle images are extracted from experimentally acquired images, placed on a synthetic background, and used to generate semi-synthetic images with consistent ground truth for evaluation of correct detection and classification by the DNN.

Oct 29, 2026

Upload by Fishy