, Google Visual Inspection AI Augments AutoML To Detect Defects In Manufacturing, The Nzuchi News Forbes

Google Visual Inspection AI Augments AutoML To Detect Defects In Manufacturing

, Google Visual Inspection AI Augments AutoML To Detect Defects In Manufacturing, The Nzuchi News Forbes

Google launched Visual Inspection AI, a new service to identify production defects in manufacturing units. The service uses the state-of-the-art computer vision models developed by the AI research teams at Google. 

Vertex AI AutoML Vision, an integral part of the managed AI platform, delivers similar capabilities. Customers can upload images and classify them based on labels before initiating a training job.  AutoML Vision generates a fully-trained model hosted in the cloud or deployed at the edge for performing inference. 

Visual Inspection AI takes AutoML Vision to the next level through its domain knowledge of the manufacturing industry. In addition, Google fine-tuned the service to speed up training and inference through domain-specific model optimization. 

According to Google, Visual Inspection AI can build accurate models with up to 300 times fewer human-labeled images than general-purpose ML platforms. This allows the solution to be deployed quickly and easily in any manufacturing setting. In addition, Visual Inspection AI customers improved accuracy in production trials by up to 10X compared with general-purpose ML approaches. And, unlike competing solutions that use simple anomaly detection, Visual Inspection AI’s deep learning allows customers to train models that detect, classify, and precisely locate multiple defect types in a single image. 

The models trained by Visual Inspection AI can be packaged as container images deployed at the edge. To accelerate inference, these containers can run on a GPU or TPU host. Unlike Vertex AI AutoML Vision that requires the same amount of labeled defects and non-defects, Visual Inspection AI requires fewer images representing possible detects in a product. The service uses image augmentation techniques to generate additional samples needed for training automatically. Through active learning, the service can quickly identify suspect examples for human review and labeling.

MORE FOR YOU

Visual Inspection AI competes with Amazon Lookout for Vision which offers similar capabilities. The key differentiating feature of Visual Inspection AI is deploying the model locally, giving more control to customers. Amazon Lookout for Vision runs the models in the cloud, which may be invoked through the command-line tool or the SDK. 

, Google Visual Inspection AI Augments AutoML To Detect Defects In Manufacturing, The Nzuchi News Forbes

Microsoft Azure Cognitive Services has an AutoML component called Custom Vision API, comparable to Vertex AI AutoML Vision and Amazon Rekognition Custom Labels. However, Microsoft is yet to deliver a verticalized AutoML vision platform targeting the manufacturing industry vertical. 

One key thing to note is that Visual Inspection AI is not available in a self-service form. The service is not integrated with Google Cloud Console, and there is no documentation or SDKs available. Google has only published a blog post and an announcement with no additional pointers to the documentation and pricing. 

Given Google’s investment in AI, it will expand AutoML Vision to other industry verticals such as healthcare and pharmaceuticals to provide customized computer vision services.

More Stories
The Southeastern Conference Will No Longer Be The SEC If Its Expansion Continues