MRTech Image Flow SDK or MRTech IFF SDK is a proprietary cross-platform library for high-performance image processing. The SDK is based on Image Flow Framework (IFF) which MRTech SK company has been developing for years.

The main purpose of IFF SDK is to deliver images to the application, the user algorithm, as efficiently as possible. MRTech team believes that with the help of IFF SDK, the user can generally achieve maximum performance for a chosen configuration of image processing system.

IFF SDK basic features

  • IFF SDK contains a description language of image processing pipeline which allows building image processing pipelines of any complexity, in full or in part.
  • A lot of processing modules are available in the library.
  • It can make effective use of the GPU when necessary.
  • IFF SDK has internal and external software interfaces for:
    • Export/Import images,
    • Use of third-party processing modules, libraries, OpenCV and AI primitives,
    • Processing and input/output control.
  • The SDK supports many machine vision cameras, such as:
    • XIMEA, Basler, other vendor cameras,
    • MIPI cameras.

Advantages of using

  • IFF SDK provides production quality code that has been proven in many projects.
  • The software focuses on high performance, low overhead, low latency.
  • The structure of IFF SDK allows for easy customization of the final solution.
  • If necessary MRTech provides support and assistance to the user with implementation.
  • The last also includes expertise, advice on equipment selection and hardware configuration.

Use cases

Processing modules:

  • Raw data processing (unpacking, split, merge, 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, sharpening
  • Image compositing
  • Other customized modules

Control interfaces:

  • JSON config file
  • HTTP REST server
  • SDK C interface

Compression and decompression:

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

Data source modules:

  • Acquisition of images from
    • XIMEA,
    • Basler,
    • MIPI and other cameras
  • Test pattern generation
  • Reading from files

Output modules:

  • OpenGL renderer
  • Raw / TIFF / DNG (including DCP) file writer
  • RTSP / RTP streaming