AT logo - The Machine Vision Specialists  
  Cameras  
  Frame Grabbers  
  Lenses  
  Lighting  
  Software  
  Vision Systems  
  Accessories  
  News  
  Careers  
  Contact AT  
  Site Map  
Request product information

Contact
info@adeptturnkey.com.au

Perth:
(08) 9242 5411

Sydney:
(02) 9905 5551

Melbourne:
(03) 9384 1775


Defence Recognised Supplier Scheme Logo

 

Special Tools for demanding imaging applications

Machine Vision Software

Xcelera-CL LX1 frame grabberIncreased production speed, more complex tasks and high-resolution, high-speed cameras require more and more computing power for imaging applications.


The Common Vision Blox GPU processing tool is the perfect solution


GPU and CPU - Perfect teamwork maximises performance. In order to increase system performance it is mandatory to use optimised algorithms and to distribute computing between more processors or processor cores. Since its first release Common Vision Blox supports multi-threading and multiprocessor systems and, thanks to its open architecture, offers the developer full control of how the application distributes its processor load.

spaAs graphics cards are designed especially for image processing and have gone through tremendous improvements to follow the requirements of the gaming industry, the logical consequence is to use the GPU also for imaging.

Thanks to their vast use in the mass market, GPUs are available at an incredible price/performance ratio and are therefore the perfect solution for many tasks in imaging applications. In addition the graphics processors out-perform other imaging acceleration methods in many technical aspects, even compared with the fastest available FPGAs. The Common Vision Blox GPU processing tool is the perfect solution, enabling the power of graphics processing units (GPUs) to be used for imaging applications:
spac-- Use the graphics card as additional independent CPU
spac-- Works asynchronous to acquisition
spac-- Allows pipelining of several tasks
spac-- Based on High Level Shader Language (part of Microsoft DirectX)
spac-- Predefined modules for standard tasks
spac-- Own algorithms can be developed flexibly using HLSL
spac-- Manufacturer independent

Why use the CVB GPU Processing tool?
As both processors have their individual strengths, it is the perfect teamwork of GPU and CPU that makes the difference. In addition it has to be noted that not every algorithm is suitable for transfer to the GPU and of course the acceleration has to offset the additional transfer. The big advantage of CVB; however, is that several algorithms can be executed on the image before it is transferred back.

The CVB GPU Processing tool offers maximum performance by:
spac-- Freeing the programmer from the complex programming of DirectX
spac-- Taking care for the memory management on the host computer and the graphics board
spac-- Organising the data transfer between host and graphics board
spac-- Allowing image access before and after GPU processing
spac-- Processing image data at full camera speed
spac-- Supporting all common acquisition technologies

Maximum flexibility is secured as the GPU is programmed by uploading the HLSL code modules upon program start or even during runtime.

Skillful use of GPU image processing avoids the need for special hardware even in applications with extremely high data throughput.

The CVB GPU processing tool combines the DirectX-API and the High Level Shader Language (HLSL) compiler from Microsoft together with functions in the Common Vision Blox software library. It enables developers of imaging software to transfer images between the host and the GPU and therefore use all processors in the system in an optimum way. Functions for GPU processing can be called from a CVB application without any additional GPU programming experience.

Multiple predefined GPU functions implement tasks such as image filtering, point operations between two images, parallel processing of four monochrome images, transformations from RGB to HSI and Bayer to RGB formats, flat field corrections, rotation and scaling of images.

Experienced HLSL programmers can also implement their own code on the GPU. To optimise the image data transfer between main memory and GPU it is also possible to combine several algorithms within the graphics card by using the open programming possibilities of the HLSL language.


Adept Electronic Solutions are "The Machine Vision and Imaging Specialists" and distributor of Stemmer products in Australia and New Zealand. To find out more about any Stemmer machine vision products, please email us at: adept@adept.net.au or call us at Perth (08) 92425411 / Sydney (02) 99792599 / Melbourne (03) 95555621 or use our online contact us page.

 

 

If you like this page, please recommend and share it.

Facebook Twitter More