Recognition of non-rigid objects
Pong Chi Yuen · 1995
The framework of a computer vision system to recognize non-rigid objects is designed and developed. Constant curvature arc segments and corners which are invariant to an object's orientation and scaling, are selected as primitives of the object. To identify an unknown object based on detected primitives, a string matching technique is developed. A set of handtools is employed to evaluate the performance of the system and the results are encouraging.