MVTec Software GmbH (www.mvtec.com), the leading provider of machine vision software, will showcase the latest developments in its embedded vision portfolio in hall 4, booth 203 at embedded world in Nuremberg. The highlights: MVTec HALCON and MERLIC for embedded vision as well as HALCON 17.12, the new version of the standard machine vision software. The latter brings powerful deep learning functions to vision systems for the first time. Users can thus train their own convolutional neural networks (CNNs) for classifying image data and use pretrained networks as a basis. This saves a significant amount of time.
Visitors can also experience future trends in machine vision by attending presentations. Christoph Wagner, Product Manager Embedded Vision at MVTec, will give a talk on "Embedded Vision - Efficient Development of Applications Using Professional Vision Software" at the embedded world Conference. Johannes Behrends, Senior Application Engineer, will provide insights into the use of deep learning technologies in the Exhibitors' Forum. His presentation is titled "Utilizing Deep Learning to Classify Image Data with MVTec HALCON."
Live demonstrations: HALCON and MERLIC for embedded vision
Practical live demonstrations round out MVTec's trade show exhibit: Experts will demonstrate the use of complex deep learning algorithms on an NVIDIA Jetson TX2 embedded board. This demonstration shows the speed and performance with which objects can be classified using deep learning - even on embedded systems. It also proves that embedded hardware is an excellent choice even for complex machine vision tasks.
Another demonstration shows four embedded boards that run machine vision applications with MVTec HALCON. A linear sled moves a box with bar codes, QR codes and OCR applications from one side to the other. Each board is connected to a camera and initiates a task. Monitors show the results live and demonstrate the performance of embedded vision with HALCON.
MVTec MERLIC is also in the spotlight: A demonstration shows MERLIC running on an ADLINK smart camera. Here, the software precisely recognizes a wide range of fonts on packaging - in fractions of a second -, including expiry dates and batch numbers, using deep-learning-based OCR. Visitors will also learn how MERLIC can be used to seamlessly integrate a programmable logic controller (PLC) into vision systems and how it visualizes the remote front end on a mobile platform.