PRODUCTS
Software
Merlic 5.1 MVTec Software will launch version 5.1 of its Merlic machine vision software on 7 April. Merlic is designed so that complete machine vision applications can be developed and operated without having to write any code. Merlic 5.1 includes a new tool for optical
character recognition, based on Halcon’s Deep OCR technology. Compared to other algorithms, this holistic deep learning-based approach can localise characters much more robustly, regardless of their orientation, font type and polarity. It also requires significantly less parameter tuning. Recognition performance is further increased by automatically grouping characters. Tis means the software can identify entire words and thus reduces the chance of misinterpretation of similar-looking characters. Te tool is included in the entry-level package, Merlic 5.1 S. Merlic 5.1 also includes artificial intelligence
acceleration interface (AI²) plug-ins for the Nvidia TensorRT SDK and the Intel distribution of OpenVino toolkit. Merlic tools using deep learning functionality can thus achieve significantly faster inference times when paired with compatible hardware like Nvidia GPUs or Intel processors. By adding support for AI², Merlic will also benefit from any future plug-ins that integrate new accelerator hardware. Finally, selecting and setting the correct
camera parameters within the image source manager (ISM) is simplified in Merlic 5.1. Merlic 5 also forms a part of the new Up
Squared Pro AI vision development kit, a vision bundle designed for users to get started with machine vision and deep learning. Te bundle combines Aaeon’s Up Squared Pro system, an AI acceleration module based on the Intel Movidius Myriad X VPU, with a Basler Pulse USB 3.0 area scan camera.
www.merlic.com
BFPython API BitFlow has introduced BFPython, an application programming interface for engineers with Python expertise to acquire images from BitFlow’s broad range of frame grabbers. Te Python bindings wrap BitFlow SDK’s
configuration, acquisition, buffer management and camera control APIs. Te download also includes several Python examples that illustrate how bindings can be used. In machine vision, where prototyping is critical to understanding how a proposed imaging solution is progressing, BitFlow BFPython accelerates the building process and reduces final development costs for those experienced with Python code. BFPython includes several sub-modules that provide interfaces to access features such as Coaxpress camera control (via GenICam) and Camera Link camera control (via the CL serial API), among others. Supporting the full line of BitFlow frame
grabbers, the BitFlow SDK enables developers to bring high-speed image acquisition into machine vision applications.
www.bitflow.com
www.imveurope.com | @imveurope APRIL/MAY 2022 IMAGING AND MACHINE VISION EUROPE 33
CxSupportPackages for 3D sensors Automation Technology has updated its CxSupportPackage to give customers a kit of software programs for the firm’s 3D sensors. Te CxSupportPackage has a variety of tools customers can use to acquire 3D images, configure the 3D camera and provide 3D data in the form of range maps or point clouds, among other things. Tese tools include CxExplorer, CxDiscover,
CxShow3D and CxSDK – the latter, which contains all the libraries for communication with the 3D camera, has been significantly expanded. Te wrappers have been revised in version 2022.2, and compatibility of MultiPart support for Automation Technology’s new C6 camera series has been enhanced. In addition, CxSupportPackage includes an optimised CxExplorer tool, which is used to configure 3D sensors. Another new feature is standard file upload and download of data containers, which contain all
Lenses
Double-sided and near-infrared aspheres Edmund Optics has added Techspec double- sided aspheric lenses and Techspec near- infrared aspheres to its selection of aspheres. Aspheres are optical lens elements with at
least one non-flat, non-spherical surface. Tere are a number of benefits associated with using an asphere over a spherical lens element, such as the elimination of spherical aberration, superior focusing ability for smaller spot sizes and reduced blur, and better light collection. Aspheres are commonly used as a lighter and smaller alternative to using a series of spherical components to accomplish the same result in a lens assembly. Te Techspec double-sided, aspheric lenses
feature high numerical apertures, diameter options from 10mm to 500mm, and are excellent options ideal for light collection or for 1:1
imaging applications. Te double-convex or bi-convex design is used to reduce the number of optical elements in the optical system and consolidate the system space and weight for superior performance. Techspec near-infrared aspheric lenses are
optimised to eliminate spherical aberration in the near-infrared spectrum. Tese lenses are manufactured from S-LAH64 or N-BK7 substrates and polished through a computer numerical controlled machining process to achieve high-precision performance. While the uncoated substrate option is designed for 780nm, other coatings options are available for 350-700nm, 600-1,050nm, and 900-1,700nm. Edmund Optics offers a wide range of
different aspheres and possesses extensive asphere design, manufacturing and metrology expertise.
www.edmundoptics.eu
measurement data and sensor configuration. Tis can be used, for example, to save sensor configurations on the PC, or customer-specific configuration and calibration information on the 3D sensor. Automation Technology has also integrated
a new software function, MultiPeak, into its Cx4090HS sensor. MultiPeak is designed to give noise-free 3D laser triangulation data, even for highly reflective materials such as metal, plastic or glass.
www.automationtechnology.de
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
