Title :
Hough space based highly parallel object recognition scheme
Author :
Chethan, Grama Y. ; Siy, Pepe
Author_Institution :
Dept. of Electr. & Comput. Eng., Wayne State Univ., Detroit, MI, USA
Abstract :
Many recognition problems can be viewed as matching images to interpretations from a model. In this paper, an effective object representation scheme based on Hough space analysis is developed to form the basis of a system for recognizing geometric objects in 2D images. This representation provides for scaling, translation, and rotation invariance of the object. A highly parallel model matching technique generates object hypotheses
Keywords :
Hough transforms; image matching; object recognition; 2D images; Hough space analysis; geometric objects; highly parallel object recognition; image matching; model matching technique; object hypotheses; object representation scheme; rotation invariance; scaling invariance; translation invariance; Data mining; Data structures; Equations; Feature extraction; Geometry; Image analysis; Image recognition; Indexing; Object recognition; Shape;
Conference_Titel :
Circuits and Systems, 1993., Proceedings of the 36th Midwest Symposium on
Conference_Location :
Detroit, MI
Print_ISBN :
0-7803-1760-2
DOI :
10.1109/MWSCAS.1993.343239
