Automating Visual Inspections in Energy and Manufacturing with AI (Cloud Next '19)

Google Cloud Tech

AES is a Fortune 500 global power company, generating and distributing sustainable energy in 15 countries, owning and managing $33 billion in assets. To optimally serve their mission of accelerating a safer, greener, energy future, AES has rigorously scaled the use of drone technologies and machine learning in their wind farm operations. Using Google’s AutoML Vision, they can help automate the detection of defects, and prioritize maintenance of their high-value assets.

LG CNS is a global IT services company with USD$2.7 billion in annual sales. They supply IT solutions for the LG Group’s affiliates and other companies, and apply big data and AI to improve manufacturing processes at scale. Originally, many inspectors were required to detect defects in everything from LCD and OLED panels, to optical films and automotive fabrics. But the monotony of visual inspections led to many errors, so LG CNS built their own in-house AI solution to visually inspect products on the assembly line. This required lots of time and effort to achieve high performing machine learning models, and they experienced a shortage of highly skilled AI experts. They turned to Google’s AutoML Vision Edge to design and distribute models to the edge, and centrally control the performance of their deployed models in one integrated system. This improved accuracy and performance, and reduced the time it takes to build high quality models.

Automating Visual Inspections → https://bit.ly/2K81RTS

Watch more:
Next ’19 ML & AI Sessions here → https://bit.ly/Next19MLandAI
Next ‘19 All Sessions playlist → https://bit.ly/Next19AllSessions

Subscribe to the GCP Channel → https://bit.ly/GCloudPlatform

Speaker(s): Mandeep Waraich (Google), Nicholas Osborn (AES), Sungwook Lee (LG CNS)

Session ID: MLAI226
fullname:Mandeep Waraich; event: Google Cloud Next 2019; re_ty: Publish; product: Cloud – General; fullname: Mandeep Waraich; event: Google Cloud Next 2019;

Source