Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications

The current study aims to train and benchmark AI models tailored for the detection of microplastic in water from scattered signals. We trained two different models, the first based on a Multi-Layer Perceptron (MLP) and the second on a Gated Recurrent Unit (GRU). A Neural Architecture Search algorith...

Full description

Saved in:
Bibliographic Details
Main Authors: Alessandro Cerioli, Lorenzo Petrosino, Daniele Sasso, Clement Laroche, Tobias Piechowiak, Luca Pezzarossa, Mario Merone, Luca Vollero, Anna Sabatini
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10990265/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850185174976823296
author Alessandro Cerioli
Lorenzo Petrosino
Daniele Sasso
Clement Laroche
Tobias Piechowiak
Luca Pezzarossa
Mario Merone
Luca Vollero
Anna Sabatini
author_facet Alessandro Cerioli
Lorenzo Petrosino
Daniele Sasso
Clement Laroche
Tobias Piechowiak
Luca Pezzarossa
Mario Merone
Luca Vollero
Anna Sabatini
author_sort Alessandro Cerioli
collection DOAJ
description The current study aims to train and benchmark AI models tailored for the detection of microplastic in water from scattered signals. We trained two different models, the first based on a Multi-Layer Perceptron (MLP) and the second on a Gated Recurrent Unit (GRU). A Neural Architecture Search algorithm was used to determine the optimal configuration for each of the two models. Moreover, for deployment on edge devices, a specific custom-made compiler was designed and used. The compiler is specifically designed for TinyML applications and, therefore, for resource-constrained devices. It bypasses traditional inference engines, compiling the NNs to native C code using only standard C libraries. Our approach demonstrated better performance compared to state-of-the-art frameworks such as <monospace>ONNX Runtime</monospace>, achieving better latency, memory usage, energy consumption, and a higher portability. This highlights the potential of our method for efficient and effective microplastic detection in environmental monitoring.
format Article
id doaj-art-4fe983229f3b4944a834b307192c7aee
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-4fe983229f3b4944a834b307192c7aee2025-08-20T02:16:49ZengIEEEIEEE Access2169-35362025-01-0113909709098210.1109/ACCESS.2025.356781610990265Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML ApplicationsAlessandro Cerioli0https://orcid.org/0009-0003-3177-7381Lorenzo Petrosino1https://orcid.org/0000-0002-1572-8075Daniele Sasso2https://orcid.org/0009-0009-3137-9715Clement Laroche3https://orcid.org/0000-0003-0401-8091Tobias Piechowiak4Luca Pezzarossa5https://orcid.org/0000-0002-0863-2526Mario Merone6https://orcid.org/0000-0002-9406-2397Luca Vollero7https://orcid.org/0000-0002-6928-0157Anna Sabatini8https://orcid.org/0000-0002-6206-5366DTU Compute, Technical University of Denmark, Kongens Lyngby, DenmarkDepartment of Engineering, Research Unit of Computer Systems and Bioinformatics, Universit&#x00E0; Campus Bio-Medico di Roma, Rome, ItalyDepartment of Engineering, Research Unit of Computer Systems and Bioinformatics, Universit&#x00E0; Campus Bio-Medico di Roma, Rome, ItalyGN Audio, Ballerup, DenmarkGN Audio, Ballerup, DenmarkDTU Compute, Technical University of Denmark, Kongens Lyngby, DenmarkDepartment of Engineering, Research Unit of Intelligent Health Technologies, Universit&#x00E0; Campus Bio-Medico di Roma, Rome, ItalyDepartment of Engineering, Research Unit of Computer Systems and Bioinformatics, Universit&#x00E0; Campus Bio-Medico di Roma, Rome, ItalyDepartment of Engineering, Research Unit of Computer Systems and Bioinformatics, Universit&#x00E0; Campus Bio-Medico di Roma, Rome, ItalyThe current study aims to train and benchmark AI models tailored for the detection of microplastic in water from scattered signals. We trained two different models, the first based on a Multi-Layer Perceptron (MLP) and the second on a Gated Recurrent Unit (GRU). A Neural Architecture Search algorithm was used to determine the optimal configuration for each of the two models. Moreover, for deployment on edge devices, a specific custom-made compiler was designed and used. The compiler is specifically designed for TinyML applications and, therefore, for resource-constrained devices. It bypasses traditional inference engines, compiling the NNs to native C code using only standard C libraries. Our approach demonstrated better performance compared to state-of-the-art frameworks such as <monospace>ONNX Runtime</monospace>, achieving better latency, memory usage, energy consumption, and a higher portability. This highlights the potential of our method for efficient and effective microplastic detection in environmental monitoring.https://ieeexplore.ieee.org/document/10990265/Microplastic detectioncompilerTinyMLquantization
spellingShingle Alessandro Cerioli
Lorenzo Petrosino
Daniele Sasso
Clement Laroche
Tobias Piechowiak
Luca Pezzarossa
Mario Merone
Luca Vollero
Anna Sabatini
Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications
IEEE Access
Microplastic detection
compiler
TinyML
quantization
title Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications
title_full Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications
title_fullStr Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications
title_full_unstemmed Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications
title_short Efficient Detection of Microplastics on Edge Devices With Tailored Compiler for TinyML Applications
title_sort efficient detection of microplastics on edge devices with tailored compiler for tinyml applications
topic Microplastic detection
compiler
TinyML
quantization
url https://ieeexplore.ieee.org/document/10990265/
work_keys_str_mv AT alessandrocerioli efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT lorenzopetrosino efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT danielesasso efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT clementlaroche efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT tobiaspiechowiak efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT lucapezzarossa efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT mariomerone efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT lucavollero efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications
AT annasabatini efficientdetectionofmicroplasticsonedgedeviceswithtailoredcompilerfortinymlapplications