Sold by: Amazon Web ServicesÂ
Deployed on AWS
A computer vision-based AutoML algorithm that trains a model to identify user specified anomalies in an object.
Overview
This pre-trained defect detection model uses transfer learning to customize a model to the user's specific application. It can be trained to support a broad range of anomaly detections such as surface damage, shape defects, and missing or out-of-place components for a specific part in the user's environment. The model supports both classification and segmentation of images.
The model can supoprt a wide range of applications, but some key use case examples include: surface inspection of automotive components looking for dents, scratches, chips, or pores; assembly verification of circuit boards looking for missing or incorrect components; manufacturing inspection looking for missing, incomplete or defective solders or welds; and equipment inspection looking for debris or signs of damage on a manufacturing line. The trained model can be run in the cloud for ease of deployment, or can be packaged to run at the edge allowing for disconnected operation and very low latency inspections.
Highlights
- This solution can be trained with as few as 20 defective and 20 normal images to allow for very rapid prototyping and POC development. Model accuracy can be improved with additional training data to enable high accuracy inspections.
- Model can scale to run on a range of hardware in the cloud or at the edge, allowing customer to determine price / performance point that fits their needs. Model will run on native CPU (ARM or x86) or GPU (Nvidia) Low latency inspection of <1 sec are achievable at the edge with mainstream hardware.
- The model leverages advanced computer vision techniques, including semantic segmentation, to not only identify the presence of defects, but also precisely locate them within the image. It can be trained to function on as few as 20 defective and 20 normal images, and will improve in performance as more data is used to fine-tune the model.
Details
Unlock automation with AI agent solutions
Fast-track AI initiatives with agents, tools, and solutions from AWS Partners.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.c5.4xlarge Inference (Real-Time) Recommended | Model inference on the ml.c5.4xlarge instance type, real-time mode | $0.00 |
ml.c5.4xlarge Inference (Batch) Recommended | Model inference on the ml.c5.4xlarge instance type, batch mode | $0.00 |
ml.g4dn.2xlarge Training Recommended | Algorithm training on the ml.g4dn.2xlarge instance type | $0.00 |
ml.c5.2xlarge Inference (Real-Time) | Model inference on the ml.c5.2xlarge instance type, real-time mode | $0.00 |
ml.c5.9xlarge Inference (Real-Time) | Model inference on the ml.c5.9xlarge instance type, real-time mode | $0.00 |
ml.c5.2xlarge Inference (Batch) | Model inference on the ml.c5.2xlarge instance type, batch mode | $0.00 |
ml.c5.9xlarge Inference (Batch) | Model inference on the ml.c5.9xlarge instance type, batch mode | $0.00 |
ml.g4dn.4xlarge Training | Algorithm training on the ml.g4dn.4xlarge instance type | $0.00 |
Vendor refund policy
This product is offered for free. If there are any questions, please contact us for further clarifications.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA)Â .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker algorithm
An Amazon SageMaker algorithm is a machine learning model that requires your training data to make predictions. Use the included training algorithm to generate your unique model artifact. Then deploy the model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
Add new hyperparameter geometric_augmentation and photometric_augmentation.
Additional details
Inputs
- Summary
Supported file format: PNG and JPEG image formats Image size: Min 64x64 pixels, Max 4096x4096 pixels All images in the dataset must have the same dimensions and orientation Maximum file size for an image in an Amazon S3 bucket: 8MB Lack of labels: Images must be labeled as normal or anomaly before training. Images without labels are ignored during training.
https://github.com/aws-samples/amazon-lookout-for-vision/tree/main/computer-vision-defect-detection/cookie-dataset/test-images
https://github.com/aws-samples/amazon-lookout-for-vision/tree/main/computer-vision-defect-detection/cookie-dataset/test-images
Resources
Support
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
SoftServe's Industry 4.0 Quick Start: Computer Vision for Quality Inspection is built on computer vision and Amazon Lookout for Vision, revolutionizing manufacturing processes, perfecting defect detection, minimizing waste, and maximizing profitability. This project serves as a stepping stone toward full-scale implementation of cutting-edge technology in your manufacturing operations.
This solution provides explanations for the predictions of a user-provided image classification model on their datasets.
Learns from visual data to detect objects, patterns, outliers and irregularities. Designed to integrate easily with your existing cloud/ on-premise/ hybrid infrastructure.
Denali Advanced Integration provides a fully managed, services led, end-to-end industrial automation solution that includes automated inspection on the edge, powered by Amazon Lookout for Vision. Our solution uses machine learning to identify anomalies in real-time to improve quality, production and customer experience while eliminating the common complexities of integrating and deploying automation in manufacturing, transportation, and logistics environments. Our dedicated team of certified engineers will tailor a solution that drives outcomes that meet your business requirements. In addition, our global footprint allows us to grow with you and scale your solution from a single facility to multiple facilities across different regions.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.