Embedded World 2019
The international trade fair for embedded systems. Alongside the exhibition halls, the Embedded World conference will run a track on embedded vision on 26 February, organised in collaboration with the VDMA. Presentations include those from Arm; AllGo Systems on implementing a camera-based driver monitoring system; Intel on developing a multi-platform computer vision solution and on enhancing human vision; Allied Vision; NXP Semiconductors on machine learning for embedded vision; MVTec Software; and Synopsys on SLAM for augmented reality.
There will also be a panel discussion on embedded vision on 27 February from 10:30am to 11:30am at the Forum in hall two.
Active Silicon
Active Silicon products will be presented at the show by ADL Embedded Solutions and EKF Elektronik.
On ADL’s booth (hall 1, booth 1-554), Active Silicon’s Phoenix PCI/104e Camera Link frame grabber and FireBird Quad USB 3.0 host controller will be on display. Active Silicon frame grabbers include ActiveDMA technology, which transfers data directly to PC memory without any host CPU intervention. This gives faster data processing from Camera Link sources including digital frame capture and line scan cameras. The host controller configures each pair of its four USB 3.0 ports to use a single lane Gen2 PCI Express interface, thereby eliminating inconsistencies resulting from shared bandwidth.
On display alongside EKF’s industrial computers (1-660) will be Active Silicon’s FireBird Camera Link 3U cPCI Serial frame grabber. The board is designed for high performance and reliability, providing fast image acquisition without any CPU intervention. It uses an FPGA, DDR3 memory and a fast Gen2 PCI Express interface.
Basler
Basler (2-550) will present its new Embedded Vision Kits, which support various embedded platforms and offer answers to customer requirements for different processors and camera interfaces. These development kits and the system software provided by Basler enable customers to integrate vision into embedded applications with plug-and-play convenience.
In live demonstrations on various topics such as IOT, AI or face detection, Basler will show what is possible with Basler products in the field of embedded technology.
There will also be presentations from Basler at the exhibitor forum and at the Embedded World Conference, which will take place at the same time.
Critical Link
Critical Link (4-180) will showcase its latest board-level solutions at Embedded World. The MitySOM-A10S is the newest addition to the company’s line of industrial embedded modules. The system on module (SOM) family is designed with the Intel/Altera Arria 10 SoC, which features dual core Cortex-A9 Arms and up to 480KLE user-programmable FPGA fabric.
In addition to the processor, the module includes on-board power supplies, two DDR4 RAM memory subsystems, micro SD card, a USB 2.0 on-the-go port, and a temperature sensor. Multiple configurations are available to meet requirements for a wide range of applications in industrial, medical, broadcast, utilities, defence, and other markets.
The MitySOM-5CSx family, based on the Intel/Altera Cyclone V SoC processor, will also be on display. It will be demonstrated live as part of the company’s embedded vision development kit featuring Bcon Dart camera modules from Basler.
Entner Electronics
Entner Electronics (3-748) will showcase a number of new products for embedded vision applications. These include the UC-310 compact zoom-block camera, a software programmable camera for embedded vision with low latency and a high quality zoom lens. The camera is compatible with the VISCA protocol, and an LVDS and/or MIPI interface is available for integration with industry standard embedded computing platforms.
The UC-200 is a small form factor camera for embedded vision with a 13 megapixel sensor and integrated F2.2 lens. The camera uses a high quality digital zoom, up to 32x, and can act as a replacement for optical zoom blocks where size and weight is important. Both the UC-310 and the UC-200 run an Arm processor onto which the user can add application-specific functionality.
Entner will also release two small form factor Flex video processing modules, for engineers to build embedded vision solutions with multiple cameras. The boards use a video processor with 3D and 4K support. Moreover an onboard Arm-based processor is open to the user to integrate DSP functionality for image fusion, stereoscopic measurement, OSD generation, file management systems, and more.
Framos
Framos will present its embedded vision portfolio at the show in hall 3A, booth 749. The Framos team will provide information on the latest advances in embedded vision and imaging components, as well as practical solutions for specific industries.
The focus will be on the latest image sensors from Sony and On Semiconductor, as well as Framos’ new product line, offering a modular approach to embedded vision development with sensor modules, and module and processor adapters. Vision engineers and developers can use the modular approach and ready-to-use compositions to accelerate their time-to-market and optimise resources.
Different processors and platforms, such as Nvidia Jetson or Qualcomm's DragonBoard, will be on display to demonstrate the flexibility in design that is available to embedded vision engineers.
Framos will show its SLVS-EC Rx IP core for Xilinx FPGAs, which can be used with SLVS-EC based sensors. Century Arks will show an embedded vision system with six camera modules that was developed for Internet of Things (IoT)-based 360° devices. The modules include the Sony IMX378 sensor, a focusable lens and image stabilisation.
As a long-term Sony partner, Framos will present a wide range of image sensors for embedded vision, OLED microdisplays and Spresense development boards. The OLED microdisplays are high-quality mini displays for applications in industrial and consumer electronics, and enable extremely high image quality and reproduction. The Spresense series includes professional development boards with Sony's six-core CXD5602 microcontroller, which can be used to develop various IoT-type applications with a short time-to-market.
Imago Technologies
Imago Technologies (1-458) will show its VisionCam Prophesee, a Linux-based smart camera able to analyse fast moving scenes. The sensor inside works more like a human eye than a classical area scan sensor.
Also on display will be the VisionCam Line Scan, a complete line scan system. The VisionCam contains a fast multi-core Arm CPU and uses the Halcon image processing library.
The board version of the entry-level VisionSensor will be introduced at the show, for OEMs that want to integrate the device into any kind of system or machine.
Irida Labs
Irida Labs (3A-549) will display EVLib, a deep learning-based software library. The software consists of CNN models optimised to power AI on edge devices such as cameras, IoT devices and embedded CPUs.
The software has more than 100 learning models that are pre-trained for accuracy, speed and low power performance. The runtime models are optimised for Arm-based CPU platforms or GPU acceleration.
Library functions include: anonymiser (blurring of people in order to remove sensitive data for GDPR); people or vehicle detection and counting; soft biometrics (male or female, age group, etc); and 2D/3D object detection and tracking.
EVLib also offers specific object detection like food recognition, product catalogue recognition, and others. Pre-trained models can be extended for custom object detection or specific scenario and tracking using additional data.
Irida Labs will also have a smart parking demo on its partner booths, Arrow (4A-340) and Analog Devices (4A-641), while a tool for visually inspecting a shopping basket in order to detect and count different retail products will be shown on the booths of NXP (4A-220), Basler (2-550), and Congatec (1-358).
MVTec Software
MVTec Software (4-203) will focus on its latest software releases, Halcon 18.11 and Merlic 4, when it exhibits at the trade fair.
MVTec will demonstrate which deep learning technologies can be used in embedded vision systems. Visitors to the booth will also find out why investing in professional machine vision software pays off.
The company will display a multi-platform live demonstration, where four embedded boards will perform different tasks using Halcon and Merlic. Other platforms that are compatible with MVTec software will be showcased offline.
The HPeek benchmarking demo program will be shown using a Raspberry Pi. In another demo, MVTec will present four standard examples of deep learning machine vision technologies in parallel – semantic segmentation, object detection, classification, and optical character recognition (OCR) – all run on an Nvidia Xavier board.
A final demonstration will cover the ways in which Merlic can be used to detect and inspect different objects based on the example of medical test tubes.
On 26 February from 2:30pm to 3:00pm, Christoph Wagner, product manager for embedded vision, will give a presentation entitled: ‘Deep learning versus rule-based-configurable vision software on embedded devices’.
Vision Components
Vision Components will present its new MIPI camera modules at the trade show at the MIPI portfolio, hall three, stand 764. Vision Components' full range of embedded vision systems will be shown at the company's main booth, 2-444.
The ultra-compact boards are designed for novel applications with many cameras, such as smart city services, autonomous driving, UAVs, and medical technology. The MIPI modules can be connected to more than 20 CPU boards from different manufacturers, including all Raspberry Pi boards, all 96Boards, Nvidia TX1 and TX2, and additional boards with i.MX6/i.MX8. They are available with various high-resolution global shutter and rolling shutter sensors.
The boards are robust and designed for good heat dissipation and minimal noise. They also contain prefabricated mounting holes and precision fittings for easy handling.
