Google Cloud launches AI dedicated to quality control and inspections

The new tool aims to help companies to reduce product defects and manage costs, the company says.

Google Cloud

Image: Dongyu Xu / Getty Images

Google Cloud announced Tuesday the launched Visual Inspection AI, a new purpose-built tool to help manufacturers, consumer packaged goods companies and other businesses reduce product defects and quality control cut costs. Visual Inspection AI builds on Google Cloud's general purpose AI product, AutoML, which has been used for years in quality control operations. According to the American Society for Quality, product defects cost manufacturers billions of dollars annually.

"Google Cloud's approach to visual inspection is the roadmap most manufacturing companies are looking for," Kevin Prouty, group vice president at IDC, said in the Google press release. "Manufacturers want flexibility, scale, inherent edge-to-cloud capabilities, access to both real-time and historical data, and ease of use and maintainability. Google is one of those companies that has the potential to bring together IT, OT and an ecosystem of partners that manufacturers need to deploy AI on the shop floor at scale."

SEE: Digital transformation: A CXO's guide (free PDF) (TechRepublic)

Using Google Cloud's computer vision technology, Visual Inspection AI helps cut inspection costs by identifying defects earlier in the production process, improving production throughput, increasing yields, reducing rework, and cutting returns and repair costs, Google said.

The launch of Visual Inspection AI comes at a time when businesses of all shapes and sizes are increasingly operationalizing AI. Google Cloud's research shows that manufacturers are keen to adopt the technology post-COVID-19.

According to Google, Visual Inspection AI is able to build accurate models with as few as 10 human labeled images, a major improvement over other machine learning modules, which usually require hundreds or thousands of images. This ability, when coupled with users' ability to train their models to detect multiple defects simultaneously using the same image, improved production trials accuracy by up to 10 times compared with general-purpose ML approaches. 

Google's Cloud Visual Inspection AI:

  • Requires no computer vision or AI subject-matter expertise to use. An intuitive user interface guides engineers through all of the necessary steps to set up and use the system.

  • Automatically increases accuracy over time as it is exposed to more products.

  • Inspection models can be downloaded to machines on the factory floor and run autonomously at the edge.

  • Visual Inspection AI is fully integrated in Google Cloud's portfolio of analytics and ML/AI tools, allowing engineers to combine insights from Visual Inspection AI with other data sources to identify root causes of quality problems or to cross-reference with supplier and customer data.

Customers can leverage Google Cloud's partner ecosystem to integrate solutions within existing operations, Google said. Google Cloud partners including SOTEC, Siemens, GFT, QuantiPhi, Kyocera, Accenture and others are supporting the deployment of Visual Inspection AI.

"With the shortage of AI engineers, Visual Inspection AI is an innovative service that can be used by non-AI engineers," Masaharu Akieda, division manager, Digital Solution Division, Kyocera, said in the Google press release. "We have found that we are able to create highly accurate models with as few as 10-to-20 defective images with Visual Inspection AI. We will continue to strengthen our partnership with Google to develop solutions that will lead our customers' digital transformation projects to success." 

Also see