Defect inspection with Deep learning
Description
Deep learning-based software aids in the recognition of images, helping machines to distinguish trends and make intelligent predictions and decisions. When deployed as part of a factory-based automation setup, deep learning-based image analysis can combine flexible intuition with computerised speed and consistency to solve machine vision applications that are troublesome to maintain due to an ever-changing population.
OVST has been dynamic in our take-up of and development involving deep learning technology for defect inspection applications. Our paint defect inspection solution, designed for deployment in automotive paint shops, is able to spot defects that may have happened during the paint application process. Our deep-learning approach enables us to accurately distinguish between true and false defects, meaning that troublesome false fail results due to particulate matter or glare are a thing of the past.
Feature
- Deep-learning based continual improvement of defect classification.
- Suitable for a wide range of inspection applications outside of the automotive paint shop.
- Fully IIoT ready: Inspection trends, pass/fail rates and failure classification results are all readily available on the cloud giving you greater control over your production processes.
Application
- Car paint inspection.
- Tyre defect inspection.
- Generic surface inspection application.