Fully integrated: The new AI Vision Solution Kit with Cloud Connectivity

AWS Panorama

AWS Panorama is a ML-based software development kit (SDK) for "on-premise" image analysis. Data collected locally in this way allows predictions with high accuracy and low latency.

This is what Arndt Bake, Chief Marketing Officer, has to say about the collaboration with AWS and the new service AWS Panorama: “Basler is looking forward to continuing our technology collaborations in machine learning with AWS in 2021. We will be expanding our solution portfolio to include AWS Panorama to allow customers to develop AI-based IoT applications on an optimized vision system from the edge to the cloud. We will be integrating AWS Panorama with our AI Vision Solution Kit, reducing the complexity and need for additional expertise in embedded hardware and software components, providing developers with a new and efficient approach to rapid prototyping, and enabling them to leverage the ecosystem of AWS Panorama computer vision application providers and systems integrators."

Discover more about AWS Panorama

Amazon Lookout for Vision

Amazon Lookout for Vision is a machine learning (ML) service that spots defects and anomalies in visual representations using computer vision.

Gerrit Fischer, Head of Marketing, also sees the intensive dialog between AWS and Basler as an important step and ultimately as a valuable customer benefit: “Many enterprises primarily rely on human inspectors to identify hundreds of different types of defects. With an estimated 1 trillion manufactured objects requiring inspection every year, scaling quality without impacting productivity is an increasing challenge for customers. We’re excited to collaborate with AWS to extend Amazon Lookout for Vision to Basler’s portfolio of machine vision camera modules and edge platforms, combining Basler’s expertise in industrial vision with AWS’ investments in industrial AI.”

Discover more about Amazon Lookout for Vision

• Fully integrated embedded vision system
• Prototyping of ML-based applications
• Simpe use of AWS services
• Optimized image processing on the edge device
• Defined interface for transferring metadata to the cloud
• Supports and shortens the actual product development
• Long-term availability of all components

Hardware

• NEW Basler dart BCON for MIPI camera with a resolution of 13 MP
• Processing: Due to its particularly powerful GPU, NVIDIA's Jetson Nano enables simplified handling of complex AI tasks

Find all product details here

First, they need to define all the components required for the system:

1. Data acquisition: A data stream of images, generated by a camera;
2. Data processing: Powerful hardware to process image data;
3. Data analytics: A machine learning algorithm needed to perform the inference and
4. Connectivity: Ability to use cloud services and send meta data to the cloud (data storage, analytics).

Next, all these parts must fit together and operate seamlessly. Again, this requires extensive expert knowledge. With its new AI Vision Solution Kit, Basler enables developers to move directly into developing their IoT vision system: all the necessary hardware and software components are integrated into a ready-to-use prototyping kit.

Targeted towards data scientists and field application engineers, the AI Vision Solution Kit is an embedded vision system for easy prototyping of IoT vision solutions. Its integrated hard- and software design comprises Basler’s newly developed IoT software architecture: Basler Container Management and Cloud Connector. With the help of the cloud services Amazon SageMaker and Amazon SageMaker Neo, machine learning models can be optimized and are ready for download to the device.

There are two options for machine learning models used on the system: developers can choose either Basler’s ML models or integrate their own ML models.

1. Basler’s ML models are pre-trained ML models for object detection and people detection. These Basler ML models are optimized for the SoC and the generated image data: they work together seamlessly.
2. With the help of our step-by-step instructions, users can also integrate their own models into the architecture and utilize the cloud services mentioned above.

Click here for comprehensive step-by-step instructions on the Imaging Hub developer page.

In the next step, the compiled ML model available in the Basler cloud is deployed on the kit using Container Management and Amazon SageMaker.

Furthermore, the parameters of the 13 MP Basler dart BCON for MIPI camera can be configured. The pylon Camera Software Suite also simplifies the operability of the camera and thus supports optimal image quality. This is important as the quality of the ML inference on the kit depends not only on the training of the model, but also on the quality of the acquired data from the camera and on the power of the processing unit. After the inference is done on the kit, the resulting meta data is sent to the AWS cloud and stored using AWS IoT Core.

With the AI Vision Solution Kit, prototyping of IoT vision solutions based on embedded technology is taken to the next level. Basler also supports its customers beyond the prototyping stage and develops customized solutions based on its flexible toolbox approach of hard- and software components.