Computing with Front Propagation: Active Contour And Skeleton Models In Continuous-Time CNN

Authors:
Csaba Rekeczky;Leon O. Chua
Affiliations:
Electronics Research Laboratory, College of Engineering, University of California at Berkeley, Berkeley, CA 94720, USA;Electronics Research Laboratory, College of Engineering, University of California at Berkeley, Berkeley, CA 94720, USA
Venue:
Journal of VLSI Signal Processing Systems - Special issue on spatiotemporal signal processing with analog CNN visual microprocessors
Year:
1999

Citing 7
Cited 4

A Computational Approach to Edge Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Propagation and its failure in coupled systems of discrete excitable cells

SIAM Journal on Applied Mathematics
Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations

Journal of Computational Physics
Fundamentals of digital image processing

Fundamentals of digital image processing
Autowave principles for parallel image processing

Selcted papers from a meeting on Waves and pattern in chemical and biological media
Shape Modeling with Front Propagation: A Level Set Approach

IEEE Transactions on Pattern Analysis and Machine Intelligence
Image processing techniques for cellular neural network hardware

Image processing techniques for cellular neural network hardware

Cellular wave computers and CNN technology - a SoC architecture with xK processors and sensor arrays

ICCAD '05 Proceedings of the 2005 IEEE/ACM International conference on Computer-aided design
Cellular neural network-based object recognition with deformable grids

Machine Graphics & Vision International Journal
Hardware Implementation of Skeletonization Algorithm for Parallel Asynchronous Image Processing

Journal of Signal Processing Systems
Energy based medical imaging segmentation methods by using cellular neural networks

Proceedings of the 15th WSEAS international conference on Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a linear CNN template class is studied witha symmetric feedback matrix capable of generating trigger-waves, aspecial type of binary traveling-wave. The qualitative properties ofthese waves are examined and some simple control strategies arederived based on modifying the bias and feedback terms in a CNNtemplate. It is shown that a properly controlled wave-front can beefficiently used in segmentation, shape and structuredetection/recovery tasks. Shape is represented by the contour of anevolving front. An algorithmic framework is discussed thatincorporates bias controlled trigger-waves in tracking the activecontour of the objects during rigid and non-rigid motion. The objectskeleton (structure) is obtained as a composition of stableannihilation lines formed during the collision of triggerwave-fronts. The shortest path problem in a binary labyrinth is alsoformulated as a special type of skeletonization task and solved bycombined trigger-wave based techniques.