Abstract
The advent of Industry 4.0 has shown the tremendous transformative potential of combining artificial intelligence, cyber-physical systems and Internet of Things concepts in industrial settings. Despite this, data availability is still a major roadblock for the successful adoption of data-driven solutions, particularly concerning deep learning approaches in manufacturing. Specifically in the quality control domain, annotated defect data can often be costly, time-consuming and inefficient to obtain, potentially compromising the viability of deep learning approaches due to data scarcity. In this context, we propose a novel method for generating annotated synthetic training data for automated quality inspections of structural adhesive applications, validated in an industrial cell for automotive parts. Our approach greatly reduces the cost of training deep learning models for this task, while simultaneously improving their performance in a scarce manufacturing data context with imbalanced training sets by 3.1% (mAP@0.50). Additional results can be seen at https://ricardosperes.github.io/simulation-synth-adhesive/.
Highlights
In several sectors of the manufacturing industry, including automotive, naval and aerospace, guaranteeing the safety of the product’s end-user is a top priority
Our contributions can be summarized as follows: 1) A novel approach to augment image datasets of structural adhesive application processes; 2) An open-source simulation to generate synthetic images of structural adhesive beads; 3) A publicly available dataset of structural adhesive applications, SEE-Q, consisting in 124 manually annotated images of real adhesive beads. This dataset can be augmented with the provided simulation to reproduce the results described ; 4) Weights and configurations for models trained and validated in a real production cell to support the reproducibility of the results; The remainder of this article is structured as follows
The demonstrator is located at Introsys S.A. facilities in Castelo Branco, Portugal, a company specializing in industrial automation that operates in the international market since 2004
Summary
In several sectors of the manufacturing industry, including automotive, naval and aerospace, guaranteeing the safety of the product’s end-user is a top priority. This makes it critical to ensure that each manufactured part adheres to strict quality criteria. A crucial step of quality assurance in these sectors consists in the tests performed after final assembly. Manufacturers integrate quality inspection of parts and components along the production line. These tests are often destructive, being performed by sampling as part of statistical quality control.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
