Most vision systems for robotic applications rely entirely on the extraction of information from gray-level images. Humans, however, regularly depend on color to discriminate between objects. Therefore, the inclusion of color in a robot vision system seems a natural extension of the existing gray-level capabilities. A method for robot object recognition using a color-coding classification scheme is discussed. The scheme is based on an algebraic system in which a two-dimensional color image is represented as a polynomial of two variables. The system is then used to find the color contour of objects. In a controlled environment, such as that of the in-orbit space station, a particular class of objects can thus be quickly recognized by its color.
Color-Coded Vision Scheme for Robotics
1991
12 pages
Report
Keine Angabe
Englisch
Manned Spacecraft , Tribology , Pattern Recognition & Image Processing , Computers, Control & Information Theory , Color vision , Computer vision , Imaging techniques , Robots , Artificial intelligence , Color coding , Edge detection , Pattern recognition , Algebra , Classifications , Color , Contours , Extraction , Polynomials , Space stations
Color and Depth Sensing Sensor Technologies for Robotics and Machine Vision
Springer Verlag | 2019
|Lisbon Robotics Cluster: Vision and Goals
Springer Verlag | 2019
|