Skilligent Computer Vision System

Skilligent Computer Vision (CV) system is a software component which implements powerful object recognition and object tracking algorithms. The system is specifically designed for robotics applications including the object recognition and tracking, image stabilization, visual-based servoing, human-to-machine interaction and visual-augmented navigation.

The Skilligent Computer Vision system keeps digital object representations in an indexed structure optimized for fast searches. The software scans the video stream coming from a camera and searches for occurrences of the objects. The computer vision software is based on algorithms resistant to

• major changes in lightening,
• partial object occlusions (up to 30-80% depending on the object and lighting conditions),
• changes of angles of view (up to 30-45 degrees),
• and camera lens distortions.

The software is based on a modified Harris Corner Detector algorithm. The software scans the video stream and extracts multiple image features. Those image features are matched against a database of known objects.

The video demonstrates how a robot recognizes various objects in a room. The robot is equiped with a low-cost video camera.

The robot shown on the video had the following objects remembered in its database:

• The upper part of the entrance door (a landmark),
• A PC on a table (a landmark),
• A picture on the wall (a landmark),
• An "Ubuntu" book,
• A "Robotics and Control" book,
• A white robot on the floor.

Skilligent Computer Vision System recongized the objects from different angles - even if the objects were partially occluded

Demo Software

Skilligent provides free computer vison demonstration software to interested parties.