Users who don’t have programming expertise can leverage Pleora’s AI modules to develop and train for defect detection, segmentation, classification, OCR, and hyperspectral imaging. The plug-ins are browser-based, meaning training and labelling can happen collaboratively in a web browser. This is ideal for multiple sites and locations sharing defect data, training and models for global quality standards.
These models are automatically optimized for the Pleora AI Gateway. This significantly simplifies the workflow from training to deployment of AI models for production. In comparison, traditional AI algorithm development requires multiple time-consuming steps and dedicated coding to input images, label defects, create custom scripts from different layers of software, fine-tune training, and optimize models.
Optionally, the AI Gateway provides a plug-in architecture that allows system integrators to develop image processing and custom AI plug-ins using standard Python programming. Integrators do not need to worry about camera connectivity or management – the output can be streamed as a GigE Vision-compatible output stream.
In an inspection application, the AI Gateway intercepts the camera image feed and applies the selected plug-in skill. The gateway sends the AI processed data to the inspection application, which seamlessly receives the video as if it were still connected directly to the camera thanks to the unique ability to stream GigE Vision video and “mirror” GenICam nodes from the connected camera.