Teledyne’s New AI Software Allows Learning at Runtime
Teledyne is pleased to announce its Sapera Vision Software Edition 2021-07 is now available. Sapera Vision Software from Teledyne DALSA offers field proven image acquisition, control, image processing and artificial intelligence functions to design, develop and deploy high-performance machine vision applications.
The new upgrades to the Sapera Vision Software include enhancements to its AI training graphical tool Astrocyte and the image processing and AI libraries tool Sapera Processing.
Recommended AI News: ThycoticCentrify Announces New Executive Leaders To Support Growth Strategy
“We are excited to introduce continual learning to the Astrocyte package. This helps improve the accuracy of a classifier ‘on the field’ by training only the samples that fail, without having to retrain from scratch,” said Bruno Ménard, Software Director for Teledyne’s vision solutions group. “Astrocyte now includes a new anomaly detection called ‘pixel-level anomaly detection’. This new algorithm is much more robust and precise than the previous one. It also has the ability to generate heatmaps at runtime,” he continued.
Sapera Vision Software is ideal for applications such as surface inspection on metal plates, location and identification of hardware parts, detection and segmentation of vehicles and noise reduction on x-ray medical images.
New features in this release:
- Continual Classification: New algorithm allows to pre-train a classifier in Astrocyte and then perform further training at runtime in Sapera Processing.
- New Anomaly Detection with Output Heatmaps: New Anomaly Detection algorithm which is more robust in locating defects while providing the ability to generate output heatmaps. Heatmaps at runtime are very useful for obtaining the location and shape of defects without the need for graphical annotations at training.
- Live Acquisition for Dataset Creation: When creating dataset in Astrocyte you can now acquire live video from a camera and generate a series of files automatically prior to training. During acquisition images are prepared for training (i.e. adjusted for size and aspect ratio) before being saved to disk.