The quality inspection of industrial products is a fundamental step in large-scale production as it boosts the yield and reduces costs. Intelligent embedded platforms with built-in tiny machine learning (tinyML) algorithms and cameras can automate quality inspection; however, running complex deep learning algorithms in low-cost and low-power embedded devices is still challenging because of the limited memory and energy resources. This paper presents an innovative sensor system with three MCU-based tinyML cameras capable of automatic artifact and anomaly detection in plastic components. The system consists of a top camera responsible for identifying shape defects and two side cameras for color anomalies. Data processing is executed locally with the tinyML reducing the data transmission to a few bytes. Two state-of-the-art convolutional neural network (CNN) architectures are evaluated, namely MobileNetV2 and SqueezeNet. Results show how both architectures – with appropriate compression techniques – are suitable to be evaluated by resource-constrained microcontrollers. The networks achieve 99% classification accuracy while maintaining suitable real-time performance, respectively equal to 5 FPS and 2 FPS.

Tiny Machine Learning for High Accuracy Product Quality Inspection / Albanese, Andrea; Nardello, Matteo; Fiacco, Gianluca; Brunelli, Davide. - In: IEEE SENSORS JOURNAL. - ISSN 1530-437X. - 2023:23.2(2022), pp. 1575-1583. [10.1109/JSEN.2022.3225227]

Tiny Machine Learning for High Accuracy Product Quality Inspection

Albanese, Andrea;Nardello, Matteo;Brunelli, Davide
2022-01-01

Abstract

The quality inspection of industrial products is a fundamental step in large-scale production as it boosts the yield and reduces costs. Intelligent embedded platforms with built-in tiny machine learning (tinyML) algorithms and cameras can automate quality inspection; however, running complex deep learning algorithms in low-cost and low-power embedded devices is still challenging because of the limited memory and energy resources. This paper presents an innovative sensor system with three MCU-based tinyML cameras capable of automatic artifact and anomaly detection in plastic components. The system consists of a top camera responsible for identifying shape defects and two side cameras for color anomalies. Data processing is executed locally with the tinyML reducing the data transmission to a few bytes. Two state-of-the-art convolutional neural network (CNN) architectures are evaluated, namely MobileNetV2 and SqueezeNet. Results show how both architectures – with appropriate compression techniques – are suitable to be evaluated by resource-constrained microcontrollers. The networks achieve 99% classification accuracy while maintaining suitable real-time performance, respectively equal to 5 FPS and 2 FPS.
2022
23.2
Albanese, Andrea; Nardello, Matteo; Fiacco, Gianluca; Brunelli, Davide
Tiny Machine Learning for High Accuracy Product Quality Inspection / Albanese, Andrea; Nardello, Matteo; Fiacco, Gianluca; Brunelli, Davide. - In: IEEE SENSORS JOURNAL. - ISSN 1530-437X. - 2023:23.2(2022), pp. 1575-1583. [10.1109/JSEN.2022.3225227]
File in questo prodotto:
File Dimensione Formato  
Tiny_Machine_Learning_for_High_Accuracy_Product_Quality_Inspection.pdf

Solo gestori archivio

Tipologia: Post-print referato (Refereed author’s manuscript)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 13.11 MB
Formato Adobe PDF
13.11 MB Adobe PDF   Visualizza/Apri
postprint-2-9_compressed.pdf

accesso aperto

Tipologia: Post-print referato (Refereed author’s manuscript)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 612.51 kB
Formato Adobe PDF
612.51 kB Adobe PDF Visualizza/Apri
Tiny_Machine_Learning_for_High_Accuracy_Product_Quality_Inspection (1).pdf

Solo gestori archivio

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 2.43 MB
Formato Adobe PDF
2.43 MB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/361562
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 3
social impact