- Anomaly detection
- Media data analysis
Automatic visual inspection of ceramic balls
Detects defects and identifies defect types without mistaking patterns on good products for defects.
- Captures images while rotating the sphere, combines three images at different angles, and uses CNN to determine Pass/Fail and classify defect type.
- Visualizes the grounds for the determination to confirm that there is no error in the determination results.
- Automatically monitors decline in accuracy resulting from changes in manufacturing conditions and equipment status.
Applications
- Automatic visual inspection at manufacturing site
Benchmarks, strengths, and track record
- Introduced to Toshiba Group manufacturing site.
Inquiries
Inquiries to the Corporate Manufacturing Engineering Center of Toshiba Corporation
Please include the title “Toshiba AI Technology Catalog: Automatic visual inspection of ceramic balls” or the URL in the inquiry text.
Please note that because this technology is currently the subject of R&D activities, immediate responses to inquiries may not be possible.
References:
- Taisuke Washitani; "Technology enabling autonomous visual inspection at manufacturing sites using deep learning"; Toshiba Review, Vol.75, No.5, pp.58-59, 2020 (in Japanese). (PDF)
- Toshiba AI Technology Catalog: Automatic monitoring technology to detect decline in accuracy of AI image inspection on manufacturing lines