15 Jun 2024

New version of no-code software

The new

MVTec Software, a leading global manufacturer of machine vision software, has announced it is launching version 5.5 of MVTec MERLIC on March 20, 2024.

The company says MERLIC is a no-code software that particularly appeals to users who have little or no experience in machine vision. “With the new version, we continue our established path with MERLIC – easy process integration combined with powerful machine vision methods. That’s why MERLIC 5.5 again includes an interface that further simplifies the process integration of MERLIC. There will also be exciting new features,” reveals Ulf Schulmeyer, Product Manager MERLIC at MVTec.

For improved process integration, a new TCP socket plugin has been developed that enables text-based communication with devices that do not support complex protocols. The new technologies include the deep-learning-based method Deep Counting for counting large quantities of objects, as well as a new tool for recognising colours. Another new feature, “High Dynamic Range (HDR) imaging”, makes it possible to address difficult lighting conditions and effectively eliminate overexposed or underexposed areas in images.

MVTec says machine vision is an important automation technology that can be used in a variety of ways, for example to ensure quality assurance, increase efficiency and reduce the workload of employees. “Companies are looking closely at further automation options. Integrating machine vision into the production can be challenging for companies. To support machine vision beginners in particular, we have developed the no-code machine vision software MERLIC. With MERLIC, it is possible to create complete machine vision applications simply by drag & drop,” says Schulmeyer.

MERLIC 5.5 extends its connectivity capabilities with the introduction of a new TCP socket plug-in. This plug-in features the ability to use a highly customisable, text-based protocol. Users can now configure specific, simple ASCII messages that contain only the information relevant to their application. These messages can be easily processed by devices that do not support complex protocols such as OPC UA. The plug-in can be easily configured within MERLIC’s RTE (Runtime Environment) Setup, providing a straightforward solution for integrating MERLIC into existing systems.

The new “Count with Deep Learning” concept tool in MERLIC 5.5 allows users to count many objects efficiently and accurately while also detecting their positions. Unlike other deep-learning-based methods, this feature can be rapidly trained with minimal labelling directly in MERLIC. This streamlines the process and significantly boosts the efficiency of object counting, particularly in scenarios involving deformable materials or bulk goods. Simultaneously, this also reduces the time and cost typically associated with deep learning model training.

MERLIC 5.5 introduces a new way to handle colours. After training, the new “Recognise Colour” concept tool enables the reliable detection of colours in various conditions. Users can even further refine detection accuracy by setting specific thresholds for acceptable deviations.

Said to be ideal for a range of use cases, such as part verification or selection, verifying proper cable connections, or confirming the installation of the correct resistor, this concept tool helps streamline and enhance quality control processes.

MERLIC 5.5 includes new methods for HDR (High Dynamic Range) imaging. This feature extends the existing “Merge Images” tool, now equipped with HDR functionality, to combine images of varying exposures into a single HDR image. This enhancement ensures that high contrast differences are handled, effectively eliminating overexposed or underexposed areas in images. As a result, even the most challenging lighting conditions can be accurately represented and analysed. This opens up new possibilities for a variety of applications, including the measurement of reflective surfaces and increasing the dynamic range of an image.

Company info: MVTec Software