polito.it
Politecnico di Torino (logo)

AI-enabled AOI: a Deep Learning-based innovative approach to improve the manufacturing processes

Valerio Di Eugenio

AI-enabled AOI: a Deep Learning-based innovative approach to improve the manufacturing processes.

Rel. Barbara Caputo. Politecnico di Torino, Corso di laurea magistrale in Data Science And Engineering, 2021

[img]
Preview
PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview
Abstract:

Reducing scraps has always been a crucial target in the manufacturing industry. In this thesis, an Artificial Intelligence application, integrated with an Automatic Optical Inspection system, is proposed to achieve this specific target. In particular, an unsupervised segmentation based on differentiable feature clustering, combined with a pattern-matching algorithm, is performed on images to identify and extract the present components one by one. Then, every single component is analyzed to verify whether it meets some imposed requirements necessary to be assembled in the final product. For this purpose, either a deterministic approach based on specific features of the components or a more general solution are tested and the results compared. Finally, to make the application perform on the assembly line, a new innovative architecture designed and build completely from scratch is presented.

Relators: Barbara Caputo
Academic year: 2021/22
Publication type: Electronic
Number of Pages: 60
Subjects:
Corso di laurea: Corso di laurea magistrale in Data Science And Engineering
Classe di laurea: New organization > Master science > LM-32 - COMPUTER SYSTEMS ENGINEERING
Aziende collaboratrici: VHIT S.p.A. Bosch Group
URI: http://webthesis.biblio.polito.it/id/eprint/20564
Modify record (reserved for operators) Modify record (reserved for operators)