Translator Disclaimer
1 March 1991 Recognition and tracking of moving objects
Author Affiliations +
Proceedings Volume 1388, Mobile Robots V; (1991)
Event: Advances in Intelligent Robotics Systems, 1990, Boston, MA, United States
A sequence of image frames after processing them digitally was used to detect classify locate track and evaluate orientation and size of moving objects. The workspace area is limited and constrained to be perpendicular to the camera axis. The calibration for the real locations and real lengthes relative to the image plane ones is done taking these constraints in account. No a priori knowledge about path of motion texture and stationary or non-stationary components of the analyzed image is assumed. A sequence of processing stages to extract the moving objects separate them code thin and fill the disconnections of the extracted moving object boundary is developed. Recognition and classification of the object is performed using Walsh-Hadamard transformation of its detected boundary. Thus the amount of computations required for this purpose is reduced.
© (1991) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
El-Sayed H. El-Konyaly, Yehia M. Enab, and Hesham Soltan "Recognition and tracking of moving objects", Proc. SPIE 1388, Mobile Robots V, (1 March 1991);

Back to Top