Introducing Visual Inspection AI for discrete manufacturing processes
In production trials, customers improved accuracy by up to
10 times than general purpose ML platform
Unlike competing solutions that use simple anomaly detection: detect, classify, and precisely locate multiple defect types in a single image
Build accurate models with up to 300 times fewer human labelled images than general purpose ML platform
Integrate your production quality data with your broader industrial data in Google Cloud with our edge-to-cloud data solutions
Our objective
⬤ Build a best-in-class visual inspection solution for manufacturing quality control
⬤ Minimize the labeled data required, powered by unsupervised learning and transfer learning
⬤ Set up and enable production workload
⬤ Scale up or down easily with Google Cloud flexible infrastructure to optimize cost
Automotive
Press shop inspection (scratch, dents, cracks, staining)
Foundry engine block inspection (cracks, deformation, anomaly)
Body shop welding seam inspection
Paint shop surface inspection
Semiconductors
Wafer level anomaly and defect localization
Die crack inspection
Pre-place inspection
SoC packaging inspection
Board assembly inspection
Electronics
PCB soldering and gluing (insufficient solder, Icicle, shift, exceeding tin, etc.)
PCB missing components (screw, spring, foam, connector, shield, etc.)
Product surface check (glue spill, mesh deformation, scratches, bubbles, etc.)
Others/Industrial
Packaging and label inspection
Fabrics inspection (mesh, tear, yarn)
Metal and plastic welding seam inspection
Surface inspection
Solution Partners
