Selectin is Energid's suite of C++ tools for machine vision. The toolkit offers unprecedented 3D object recognition and tracking technology in an easy-to-use software package. With Selectin, developers can build object recognition systems simply by training the algorithms on detailed CAD models of the objects to be identified. Selectin can distinguish between types of objects, and it can determine the pose (i.e. position and orientation) of the objects in the scene. It can even track object articulation (useful for tracking everything from the blades on a pair of scissors to the length of a bolt).

The Selectin approach to C++ object recognition and machine vision is flexible and extendible, with application to many different types of imaging sensors, including gray-level, color, hyperspectral, and ladar sensors. Central to the Selectin approach is the ability to trade off run time for accuracy at every step in the algorithm.

The C++ object-oriented framework includes spatial components to analyze each sensor data set in isolation and temporal components to combine spatial-analysis results over time. The object recognition technique uses XML-based CAD models of the objects to be recognized to automatically build tailored algorithms.

Selectin Toolkit Advantages

• 3D Recognition and Tracking - tracks position, orientation, and articulation of targets from CAD models.
• Speed - fast tracking is achieved by using powerful, low-cost PC graphics cards for signal processing
• Flexibility - enabled by an XML-based scripting language. Supports color, gray level, and ladar.
• Code Reuse - control algorithms developed for one application can be used on others.

Do you have any questions or would you like more information on Selectin? Please contact us.