DocumentCode
358263
Title
Application of direction constrained and bipolar waves for pattern recognition
Author
Petrás, István ; Roska, Tamás
Author_Institution
Comput. & Autom. Inst., Hungarian Acad. of Sci., Budapest, Hungary
fYear
2000
fDate
2000
Firstpage
3
Lastpage
8
Abstract
Direction constrained and bipolar waves are introduced. Their possible applications for direction selective curvature and concavity detection as well as region segmentation are shown. A cellular neural (CNN) algorithm frame for feature-based object decomposition is presented. Algorithms are tested on the 64×64 CNNUM (CNN Universal Machine) chip
Keywords
cellular neural nets; filtering theory; image segmentation; neural chips; object recognition; CNN Universal Machine chip; bipolar waves; concavity detection; direction constrained waves; direction selective curvature detection; feature-based object decomposition; region segmentation; Application software; Cellular neural networks; Computer networks; Laboratories; Logic; Pattern recognition; Positron emission tomography; Spatiotemporal phenomena; Testing; Turing machines;
fLanguage
English
Publisher
ieee
Conference_Titel
Cellular Neural Networks and Their Applications, 2000. (CNNA 2000). Proceedings of the 2000 6th IEEE International Workshop on
Conference_Location
Catania
Print_ISBN
0-7803-6344-2
Type
conf
DOI
10.1109/CNNA.2000.876810
Filename
876810
Link To Document