Image processing

High-performance software for image processing

MRTech SK presents image processing software and integrated solutions for Machine Vision and Embedded systems. This software helps our customers to build modern, highly demanding video systems by effectively getting the images from their source and delivering them to where they need to be with maximum performance and minimum latency for selected software and hardware configuration.

MRTech SK provides its clients with processing software:

  • in form of a specially designed application based on our proprietary cross-platform Image Flow Framework or IFF and
  • as Image Flow SDK or IFF SDK which allows the user to create his own image processing application for machine vision system.

MRTech Image Flow Framework is designed to build image processing pipelines of any complexity, to control camera, processing, and output configuration, as well as to add any customized, third-party filters, libraries, primitives which needed for the vision system.

IFF software has production style, high quality code, and as far as we know from our experience gives the user the following benefits:

  • Easily customized image processing pipeline;
  • Low overhead and latency;
  • Solutions based on IFF can usually achieve maximum possible performance for a chosen hardware and software configuration.

MRTech IFF software is compatible with all major hardware platforms common on the machine vision market, works with Linux, Windows, MacOS operating systems, when necessary and possible uses NVIDIA CUDA computing platform.

Moreover, MRTech offers its customers the services of software and system support, processing customization, new and custom processing modules implementation.

A lot of processing functions and modules are available in the MRTech IFF, some of them are described below.

Base image processing

  • RAW data processing (unpacking, split, merge, denoise, etc.)
  • Dark-frame subtraction
  • Flat Field Correction (FFC)
  • White balance, auto white balance
  • LUT
  • Debayer (different algorithms)
  • Spatial denoiser
  • Color transforms (RGB, HSV, YCbCr)
  • Color correction
  • Image affine transforms
  • Crop, resize, sharp
  • Image compositing
  • DNG processing including DCP profile
  • Other customized modules

Compression and decompression

  • JPEG codec
  • RAW Bayer codec
  • JPEG2000 codec
  • H.264 / H.265 codec

Data source modules

  • Acquisition from cameras of various brands, or from custom cameras
  • Test pattern generation
  • Reading from files

Output modules

  • OpenGL renderer
  • Async disk writer
  • RTSP / RTP / HTTP streaming

How it all works if you would like to look and use our software

 

  • You request us and describe your requirements;
  • If necessary, we provide a demo to you;
  • As a rule, a mutual NDA is signed, and a specification or a proposal agreed upon;
  • We design the software, check and test it with the right equipment;
  • When necessary, we develop architecture, assemble, and test prototype or production system;
  • We deliver software to you with documentation and with or without hardware as agreed;
  • We consult you and maintain our software for you.

Please contact us for more information.