​
New Features

  • Launched a new Support Center that includes new and improved documentation.
  • Added Classification to the improved workflow that allows for faster Model creation and training. (Previously this workflow only supported Object Detection and Segmentation.) (Beta feature*)
  • Added a new Project Type called Anomaly Detection. Anomaly Detection is used to identify any deviations from the norm and can be trained entirely on images without the Class you’re looking for. (Beta feature*)

*Beta features are deactivated by default. If you would like a Beta feature to be activated, please contact .

​
Improvements

  • Improved performance for datasets of 50 or fewer images. This development also increased the Model training speed by about 10%.
  • Increased memory limitations from 2GB to 3GB for Model training.

​
Bug Fixes

  • Fixed issue where the LandingLens user interface randomly closed and restarted.
  • Fixed issue where Inspection Points in LandingEdge didn’t save properly.
  • Fixed issue where Inspection Start times in LandingEdge reverted to 5 seconds if the interval was set to less than 1 second.
  • Fixed issue where BMP images didn’t load properly in LandingEdge input folders.
  • Fixed issue where the zoom levels were the same for all images when running inference in LandingEdge.
  • Fixed issue where Model names in LandingEdge weren’t being saved.