Imagery creation based on autonomic system for finite element by using fully convolutional network

Gentaro Miyaki, Komei Tanaka, Hiroshi Hasegawa

Research output: Contribution to journalConference articlepeer-review


For applications in Industry 4.0, a system that can analyze the deformation and stress of a target object from an image acquired by a smartphone or tablet is proposed in this paper. The process for the proposed system is more convenient for users than creating a computer-aided design model. The target objects include bridges and plant equipment, and the proposed process aims to facilitate maintenance inspections. To acquire results from this system, a fully convolutional network is employed to extract the target object from the obtained image, and density-based topology optimization is applied to produce a finite element model, which is imported to commercial software. Artificial intelligence image processing is adopted to generate the output of the finite element model for the target object. In this work, numerical examples demonstrate that the final model for the target object is accurate and appropriate for finite element deformation and stress analysis.

Original languageEnglish
Pages (from-to)383-386
Number of pages4
JournalProcedia Manufacturing
Publication statusPublished - 2020
Event1st International Conference on Industry 4.0 and Smart Manufacturing, ISM 2019 - Rende (CS), Italy
Duration: 2019 Nov 202019 Nov 22


  • Autonomic Generation
  • Finite Element Model
  • Fully Convolutional Network
  • Image Processing

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Artificial Intelligence


Dive into the research topics of 'Imagery creation based on autonomic system for finite element by using fully convolutional network'. Together they form a unique fingerprint.

Cite this