At ITE 2018 (International Technical Exhibition on Image Technology and Equipment) in Yokohama (Japan) from December 5th to 7th we will present our hardware and software products for machine vision in hall D, booth 1 together with our Japanese partner LINX Corp. Focus is on Deep Learning, embedded Vision and Computational Imaging for a closer integration of image processing into factory automation by technologies like OPC UA Vision, Industry 4.0 and Industrial Internet of Things. Our image processing solutions significantly accelerate the development of image processing applications and the time-to-market of complete visual inspection systems.

Logo ITE

The use of image processing on programmable image acquisition and processing boards or (embedded) components such as cameras and vision sensors with FPGA processors guarantees high speeds and large data throughput in production environments with very low latencies and low thermal power. This benefits industries such as the electronics and semiconductor industries, robotics and autonomous driving for applications such as quality control, inline inspection, condition monitoring, predictive maintenance, machine control, human-machine interaction and pick & place. For complex inspection tasks such as feature classification and defect detection, we will be demonstrating at our booth a deep learning application with a high data rate of over 200 MB/s and real-time output of inspection results.

Overview Product Highlights

Our image acquisition and processing boards (frame grabbers) deliver high-speed results in the production cycle and centrally control peripherals such as lighting, cameras, actuators and sensors in the factory floor.

  • High-speed Deep Learning solution: For especially complex applications we will show the programmable microEnable 5 marathon VCLx Camera Link frame grabber, a very fast board which runs our VisualApplets software
  • High-speed automation solution: LightBridge 2 is a small external image acquisition and processing device that is integrated into the production process and transmits the result images via optical data connections (ThunderboltTM 2 technology)
  • High-speed transfer solution: The CoaXPress microEnable 5 marathon frame grabber family has been widely expanded. The new cards support different topologies from single to dual to quad configurations (from CXP1 to CXP6).

Our hard- and software control the entire image processing system

Deep Learning as CNN operator in a dataflow diagram

Image processing applications in the 2D, 3D, multi- and hyperspectral range as well as deep learning with neural networks (CNN) can be implemented in a short time with our graphical development environment VisualApplets via data flow diagrams on FPGA processors, even on embedded devices.

  • High-speed application solution: For deep learning applications, VisualApplets adopts pre-trained deep neural networks and their weights and integrates them as a CNN operator into the application design, which can be combined with additional image preprocessing, postprocessing, and signal processing. Users simply continue to use their existing image processing system.
  • High-speed development solution: With VisualApplets and the extension VisualApplets Embedder (embedded) vision devices can be programmed easily and quickly with decentralized image processing intelligence (edge computing) also for deep learning. The VisualApplets Expert extension makes it easier for users to transfer VHDL code to VisualApplets as generic operators. Thanks to the new Tcl script console, design processes can be automated.

ITE 2018
Hall D, Booth 1, in front of our partner LINX Corp.
5th-7th December 2018 in Yokohama, Japan

Distributor: LINX Corp.

Complete lecture program ITE 2018

Next trade fair participations in Asia
Vision China, Shanghai, 20th-22nd March 2019
Korea Vision Show, Seoul, 27th-29th March 2019

Our Sales Manager APAC Rabia Muhammad and Marketing Manager Emi Eguchi are looking forward to your visit at our booth.