Posted On: Jul 29, 2022
Amazon Lookout for Vision now provides anomaly localization through semantic segmentation. You can use Lookout for Vision’s segmentation models to identify the locations on an image where different types of anomalies (such as a scratch, dent, or tear) are present, the label of the anomaly and the size of the anomaly, which you can then use to make decisions such as classify, grade, bin product and ship, rework, or scrap the part. You can deploy the trained semantic segmentation models for inference in the AWS Cloud via the AWS SDK or CLI. You can also deploy them to an edge hardware device of your choice and run inference locally on the device.
Amazon Lookout for Vision models can now be deployed to run inference locally on an x86 compute platform running Linux with just a CPU, without needing a GPU accelerator. This will help you save on hardware cost, enable inference in low power mode (such as for use in mobile rovers), and enable a large selection of x86 based edge appliances to choose from. This is in addition to deploying on any NVIDIA Jetson edge appliance or x86 compute platform running Linux with an NVIDIA GPU accelerator. You can use AWS IoT Greengrass to deploy and manage your edge compatible customized models on your fleet of devices. AWS IoT Greengrass is an open-source edge runtime and cloud service for building, deploying, and managing device software.
With Amazon Lookout for Vision, you can use computer vision to automate visual inspection during tasks like quality control and defect assessment, precisely locating different types of defects. You can proactively identify the location of defects such as damaged parts (dents, scratches, or poor welding), missing product components, or defects with repeating patterns, to improve efficiency of quality control.
Amazon Lookout for Vision is available directly via the AWS console as well as through supporting partners to help customers embed computer vision into existing operating systems within their facilities. Amazon Lookout for Vision is available in US East (Ohio), US East (N. Virginia), US West (Oregon), Europe (Ireland), Europe (Frankfurt), Asia Pacific (Tokyo), and Asia Pacific (Seoul), with availability in additional regions in the coming months. To learn more and get started, visit the Amazon Lookout for Vision product page.