This work was targeting the identification of tiny visual defects which were present on mechanical components in die-cast aluminum, just produced by a foundry. From those, an image data-set has been created composed by 500 colored pictures, 2048x850 pixel resolution each. The tools to perform the analysis were based on Artificial Neural Network (ANN) to achieve accurate defects detection. Moreover we were interested to map resulting ANNs topologies on micro-controllers MCUs as well as investigating architectures for efficient mapping. At first we have designed an ANN without tight implementation constraints, memory and computational, achieving 86% accuracy. Careful analysis of memory and computational needs, by using X-CUBE-AI tool, concluded it required an embedded RAM memory of 786 KB combining 64x64 input and activation buffers. To lower that number we explored an ANN implementation by removing the residual connections between convolutional layers featuring stride 2. That version achieved an accuracy of 75% in 100KB of embedded RAM.

(2021). Tiny defects identification of mechanical components in die-cast aluminum using artificial neural networks for micro-controllers . Retrieved from https://hdl.handle.net/10446/234069

Tiny defects identification of mechanical components in die-cast aluminum using artificial neural networks for micro-controllers

Previdi, Fabio;
2021-01-01

Abstract

This work was targeting the identification of tiny visual defects which were present on mechanical components in die-cast aluminum, just produced by a foundry. From those, an image data-set has been created composed by 500 colored pictures, 2048x850 pixel resolution each. The tools to perform the analysis were based on Artificial Neural Network (ANN) to achieve accurate defects detection. Moreover we were interested to map resulting ANNs topologies on micro-controllers MCUs as well as investigating architectures for efficient mapping. At first we have designed an ANN without tight implementation constraints, memory and computational, achieving 86% accuracy. Careful analysis of memory and computational needs, by using X-CUBE-AI tool, concluded it required an embedded RAM memory of 786 KB combining 64x64 input and activation buffers. To lower that number we explored an ANN implementation by removing the residual connections between convolutional layers featuring stride 2. That version achieved an accuracy of 75% in 100KB of embedded RAM.
2021
Pau, Danilo; Previdi, Fabio; Rota, Emanuele
File allegato/i alla scheda:
File Dimensione del file Formato  
2021 IEEE ICCE - Defect deep learning ST Micro.pdf

Solo gestori di archivio

Versione: publisher's version - versione editoriale
Licenza: Licenza default Aisberg
Dimensione del file 2.41 MB
Formato Adobe PDF
2.41 MB Adobe PDF   Visualizza/Apri
Pubblicazioni consigliate

Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10446/234069
Citazioni
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 0
social impact