MVTec Software has launched the latest version of its machine vision software Merlic, which features a new deep learning object detection tool.
The “Find Object” tool can locate trained object classes and seperate those that are touching or partially overlapping. This allows the objects to be counted.
Merlic version 5.3 builds on its predecesor 5.0, which features deep learning technologies such as anomaly detection and classification, deep OCR, and global context anomaly detection. With the new deep learning tool, users can label and train networks, and then load them into Merlic at the click of a button, helping to solve complex machine vision applications.
The software is suitable for beginners, according to the firm, as no programming knowledge is required to implement it.
Merlic 5.3 provides a visual tool flow to improve the clarity of workflows, alongside a concise startup dialogue with thumbnails for easy access to functions and machine vision applications. This allows users to get an overview of their most recently opened MVApps. Additionally, all standard examples are clearly displayed and introductory material for new users has been provided.
Thanks to a new plug-in that supports the Mitsubishi Electric iQR, iQF, L- and Q-series, Merlic 5.3 can also communicate directly with the widespread Mitsubishi Electric PLC via the MC/SLMP protocol.
"All features are easy-to-use,” says Dr Maximilian Lückenhaus, Director of Marketing and Business Development at MVTec Software.
“This means that even complex machine vision applications can be implemented with our software without any programming knowledge."
MVTec has also added training features to Merlic 5.3 that can be used in the frontend during runtime, which allow users to train new matching models or code reading parameters.
This means that the end customer can perform training for other products directly on the production line in the Merlic frontend, which increases flexibility and application possibilities, according to the firm.
Merlic 5.3 is now available.
