MRTech IFF SDK
About
MRTech IFF SDK is a powerful cross-platform toolkit that facilitates the development of high-performance machine vision and image processing applications. The core of IFF SDK is Image Flow Framework which was initially developed by MRTech in 2016 and has been continuously improved and used in many projects ever since.
The main feature of IFF SDK is the delivery of images to the customer’s code with maximum efficiency. We believe that with the IFF SDK our customers can achieve maximum performance for the chosen configuration of the image processing system.
Basic features
- A text-based pipeline configuration format that allows you to create a highly complex image processing workflow
- A wide range of processing modules available to the customer and working out-of-the-box
- Hardware accelerated processing on Nvidia GPUs
- Ability to export and import images from the SDK pipeline to the customer application
- Controlling the parameters of processing modules at runtime
- Easy integration with OpenCV, third-party processing libraries, and custom processing modules
- Supporting a large number of machine vision cameras, such as:
- XIMEA, Basler, other vendor cameras
- MIPI cameras
Advantages
- Production-ready high quality code, successfully used in many projects
- High performance image processing with low latency and low overhead
- SDK architecture that makes it easy to develop and customize the target application
- If required, we can provide the customer with support and assistance in the implementation of the project, including project study, recommendations on hardware selection and system architecture, code review
Application examples
- Low latency video processing for teleoperation system for ETH Zurich, Robotic Systems Lab
- Image processing and video delivery for augmented reality system prototype for Command Sight, Inc
- Solutions for NVIDIA Jetson platform
Processing modules
- Dark-frame Subtraction
- Flat Field Correction (FFC)
- White Balance
- Gamma Correction
- LUT
- Image Demosaicing (HQLI, L7, DFPD, MG algorithms)
- Histogram Calculation
- Spatial Denoiser
- Colorspace transforms (RGB, HSV, YCbCr)
- Color Correction
- Image Crop
- Image Resize
- Image Overlay
Control interfaces
- JSON Configuration files
- HTTP JSON API
- C-library SDK Interface
Compression and decompression
- JPEG
- JPEG2000
- H.264
- H.265
Input capabilities
- Image Acquisition From Camera (XIMEA, Basler, MIPI, etc.)
- RTSP Video Source
- File Data Input
Output capabilities
- On-screen Render via OpenGL
- Raw / TIFF / DNG (including DCP) file data output
- RTSP Video Stream
Custom modules
- See Services