Camera Module Surface Inspection
Ensure that camera lenses are free of surface defects
Prodotti correlati
Software di visione basato sul deep learning con tutta la potenza di In-Sight ViDi
A breakthrough in complex inspection, part location, classification, and OCR
Before the camera module is installed in a mobile device, its surface must be inspected to ensure that there are no foreign materials, scratches, smearing or dust on the lens.
Different defects vary widely in appearance. A smudge from a fingerprint looks quite different from dust particles trapped under the lens coating, and neither looks anything like a scratch on the glass surface. In addition, the lens’s reflective surface and refracted images from parts underneath the lens that can appear as unwanted anomalies even though they are actually not. Distinguishing these background anomalies from true defects often requires manual inspection, which is slow, costly, and inconsistent. Yet traditional rule-based machine vision systems cannot be easily programmed to consistently identify such a wide span of defects.
Cognex Deep Learning’s defect detection tool is trained on a wide selection of defect-free lenses to learn the full variation of normal parts. In Unsupervised mode, it then scans through a sequence of lenses and flags any that are outside of the acceptable range, while minimizing false positives.
Lens defects tend to have certain characteristics stemming from specific causes, such as contamination by dust and other particles, smearing from oil or fingerprints, and misalignment of internal components. Users who need to identify the specific type of defect, or precisely measure the defect’s size, can use Supervised mode. In this mode, the user trains the system on a combination of good and bad parts, explicitly highlighting the defect regions and labeling them with the type of defect: scratch, stain, contamination, or others.
This knowledge can be used for upstream process control. A certain type of scratch might be caused by a misaligned machine, or a fiber might be deposited due to poor airflow in a manufacturing process. By identifying the root cause of the problem, manufacturers can quickly take corrective action and minimize the creation of more bad parts.
