Computer Vision

Key Area of Our Scientific Research

Softarex Technologies is one of a few companies conducting ongoing research in the fields of Computer Vision, Pattern Recognition Systems, Robotics Management Systems, Multimedia Data Transfer Systems, Real-time Video Stream Recognition Systems and Interactive 3D Scene Modeling Systems.

Find the Right Solution

Modern information technologies and computer vision systems have opened up new ways for processing visual information received from video input devices, such as video cameras, digital photo cameras and scanners. The success of creating efficient pattern recognition systems is in the development of effective algorithms for images processing and their subsequent implementation in software.

Our advantages:

  • Highly qualified specialists in the Pattern Recognition field
  • Partnerships with universities in joint, scientific-based research projects
  • Our specialists have published more than 50 scientific papers in the Pattern Recognition field
  • Our specialists have been working in this area for over 10 years
  • Our consultants have been working in this area for over 30 years
  • Strong focus on implementation of pattern recognition algorithms in real-life applications

Theoretical Background

In our work, we use different methods for pattern recognition and images processing. Here are the most commonly used of them:

  • Images Normalization (automatic geometric transformation compensation including projection transformation). This is a set of methods developed by our team.
  • Tracking Normalization – methods which allow tracking moving objects and different information for object management and recognition. This method was also developed by our team.
  • Neural Networks
  • Segmentation algorithms including those developed by our team
  • Moving detection based on differences in images
  • Objects recognition based on neural networks, correlations, partial correlations and points of interest
  • Histograms analysis
  • Cluster analysis
  • Differential equations, integrals, mathematical morphology, numerical methods including our own solutions used for images processing
  • Skeleton Building Methods developed by our team

Static Images Analysis

Objects analysis and identification, including the number, size and shape of objects. This is applicable for:

  • Counting the number of microbes in water, simultaneously identifying their size, color and type and then checking for similar objects in a comparative database as well as performing renewal of that database;
  • The analysis of Clinker quality with the recognition of the object’s structure and identification of its form and size;
  • Studying the structure and make-up analysis of blood by identifying and counting all immersed objects;
  • Identification of objects in images;
  • Raster image tracing and conversion into vector images;
  • Algorithms for Optical Character Recognition (OCR).

System Concept

  • Scene editor with all the required scene construction features
  • Scripting language and editor to describe scenarios for objects’ behavior
  • Physical platform with all the necessary functionality for modeling mechanic processes and interactions between objects
  • Ability to scale the system by adding modules, designed by other developers
  • Import of models from .X and .fbx file formats and graphic file formats
  • Ability to record and reproduce user’s actions
  • Notifications about user’s mistakes while working on a scene
  • Works on Windows 7, XNA 3.1

Dynamic Images Analysis

The following tasks can be resolved with dynamic images analysis:

  • Tracking of moving objects (with emergency alarm if necessary);
  • Vehicle Identification Number plate tracking and recognition in real time under the conditions of noise, different luminance and local obstacles;
  • Faces recognition;
  • Identification of objects in video streams;
  • Mobile management and control systems for robotics (we are conducting research in this area).

SceneStudio: Visualization and Modeling of Interactive 3D Scenes

  • The necessity to have systems capable of visualizing various situations with the purpose of their visual analysis
  • Increase the effectiveness of specialists’ training
  • Analysis of possible consequences of an accident at the objects and system design stage
  • Analysis of possible consequences of an accident on the already existing objects and systems
  • Reduce efforts and save time when developing visualization systems and modeling interactive scenes

Where can SceneStudio be used?

  • Development of simulators/training programs with the use of interactive 3D graphics
  • Visualization and modeling of different processes allowing users to take part in these processes
  • Systems for modeling and analysis of emergency situations