FPGAs garnered much of the attention during The Vision Show that took place May 3-5 in Boston. Other products on exhibit ranged from industrial robots to lenses.
FPGAs were the topic of two presentations at the show. Paul Scardino, senior vision application consultant for Baumer, discussed FPGAs as key components of the company’s LX cameras with VisualApplets technology. The term VisualApplets refers to the graphical development environment for FPGA programming from Silicon Software, which also participated in the show. With FPGA preprocessing, Scardino said, the amount of image data transmitted can be reduced, making it possible, for example, to employ GigE Vision instead of Camera Link interconnect.
And Jim Blasius, solutions architect for measurement and automation at ADLINK Technology, described his company’s Neon smart camera, which includes an Atom E3845 quad-core processor and FPGA coprocessing. The FPGA can handle tasks like shading correction without using CPU resources, he said. ADLINK also exhibited its four-channel GigE EOS compact vision system with 6th-generation Intel Core processors and the new Matrix MVP series of embedded computers targeting single-purpose vision applications.
National Instruments highlighted its NI Compact Vision Systems, which combine industrial camera connectivity, open communication, and FPGA-based I/O in a small form factor. The systems can acquire and process images in real time from multiple cameras. Powered by Intel Atom processors, the systems are compatible with GigE Vision or USB3 Vision cameras. They include real-time display and industrial communication ports and industrial digital I/O that you can customize using the LabVIEW FPGA Module.
NI also presented its CompactRIO controllers, which use Intel Atom processors and Xilinx Kintex-7 FPGAs, offering connectivity to GigE Vision cameras and USB3 Vision cameras. The processor handles network communication, data logging, control, and processing with the deterministic and reliable NI Linux Real-Time OS. The user-programmable FPGA can implement custom hardware for high-speed control, inline data processing, or complex timing and triggering. In addition, the FPGA can be used for image coprocessing, which reduces latency and processing time so cameras can provide an input for closed-loop control applications.
Matrox Imaging highlighted its Matrox Design Assistant 4—a hardware-independent integrated development environment that lets users easily create an application flowchart and HMI. The software works with the Matrox Iris GTR smart camera, which measures 75 mm x 75 mm x 54 mm. The cameras use ON Semiconductor’s PYTHON CMOS image sensors with high readout rates and an Intel Celeron dual-core embedded processor.
XIMEA exhibited its new Model MQ022HG-IM-LS150-VISNIR camera, a member of the company’s hyperspectral xiSpec Series. While conventional color cameras use a mosaic pattern of red, green, and blue (RGB) color filters on top of the pixel matrix, hyperspectral cameras replace the RGB pattern with many different wavelength filters. The features of this new xiSpec multi-linescan HSI camera include 150 HSI bands between 470 and 900 nm, covering the visual and NIR spectrum, and high-speed multi-lineÂscan capability with up to 1,020 lines/s. The cameras are USB3 Vision compliant and include Windows, Linux, and MacOS drivers; a software development kit (SDK); and a USB 3.0 micro-B connector with a screw-lock.
Microscan introduced its latest technology platform, combining its smallest-ever imaging devices with a new release of AutoVISION machine-vision software. The result is a fully capable barcode reading suite consisting of four unique smart cameras, which will adopt the complete machine-vision toolset of AutoVISION for the company’s “Auto ID+” applications.
Edmund Optics showcased its new TECHSPEC UC Series of fixed-focal-length lenses. Designed to work with all smaller-format camera sensors and built specifically for machine-vision working distances and resolution requirements, these compact lenses are suitable for a range of applications including factory automation, inspection, and biomedical instrumentation.
Teledyne DALSA featured its newest cameras and vision solutions, including the Linea line-scan cameras as well as next-generation NIR, LWIR, and X-ray cameras, including the Piranha4 multispectral camera, the Calibir uncooled LWIR camera, and the Rad-icon flat-panel X-ray detector for nondestructive testing. The company also announced the addition of eight new models to its Genie Nano Series GigE Vision cameras, including four monochrome and four color models that incorporate Sony Pregius image sensors. The addition of these eight new models brings the total number of cameras in the series to 27, with more models planned.
The new models, the M2450, C2450, M2050, C2050, M2420, C2420, M2020, and C2020, are designed for industrial-imaging applications and are capable of data transfer rates at two or even three times the standard GigE Vision rates. The Genie Nano series also takes advantage of the Sapera LT SDK and the vendor’s Trigger-to-Image-Reliability framework for full system-level monitoring, control, and diagnostics from image capture through transfer to host memory.
Robotics also was a key focus of the show. ABB featured two YuMi dual-arm, collaborative robot demonstrations: a vision-enabled, small parts assembly application and an interactive, hands-on demo allowing attendees to program YuMi using ABB’s lead-through programming technology.
