Vision 2021: Merlic machine vision software gets deep learning update
The improved camera handling is based on the Image Source Manager (ISM). This separates the image acquisition source from the image processing part so that machine vision applications can be easily copied and executed between different systems without adjusting the camera settings. The graphical user interface enables convenient management and configuration of the image sources.
Two Deep Learning functions
In addition, with the release of the new version, MVTec's latest deep learning functions are now also available in Merlic 5. Among other things, the machine vision software includes deep-learning-based anomaly detection. Anomaly detection is especially useful for applications with no or few "bad images". For training purposes, just a few "good images" are sufficient to locate defects of any kind accurately and precisely.
The second Deep Learning feature integrated in Merlic 5 is the "Classify Image" tool. With its help, images can be assigned to trained classes on the basis of Deep Learning algorithms. The technology is particularly suitable for classifying defects or objects with high variance in shape and appearance, such as natural products.