Showing papers by "Wen-Hsiang Tsai published in 1992"
••
TL;DR: A new parallel thinning algorithm for binary images employs template matching to remove the edge points of an object shape in a binary image iteratively and obtains skeletons with the properties of perfectly 8-connected, noise-insensitive, and topologically equivalent to the original object shape without excessive erosion.
55 citations
••
TL;DR: A mapping method that makes the Hopfield neural network perform the relaxation process is proposed and it is shown that the neural network technology can be easily adapted to solve the many problems which have already been solved by the relaxed process.
52 citations
••
TL;DR: Two new types of shape-specific points, called fold-Invariant centroid (FIC) and fold-invariant radius weighted mean (FIRWM), are introduced for detecting the orientations of rotationally symmetric shapes.
32 citations
•
21 Jan 1992TL;DR: In this article, a neural network comprising an array of neurons interconnected by synapses (i.e., weighted transmission links) is utilized to carry out a probabilistic relaxation process.
Abstract: In accordance with the present invention, a neural network comprising an array of neurons (i.e. processing nodes) interconnected by synapses (i.e. weighted transmission links) is utilized to carry out a probabilistic relaxation process. The inventive neural network is especially suited for carrying out a variety of image processing tasks such as thresholding.
12 citations
••
TL;DR: A single-layer recurrent neural network is proposed to perform thinning of binary images by iteratively removing the contour points of an object shape by template matching.
Abstract: A single-layer recurrent neural network is proposed to perform thinning of binary images. This network iteratively removes the contour points of an object shape by template matching. The set of templates is specially designed for a one-pass parallel thinning algorithm. The proposed neural network produce the same results as the algorithm. Neurons in the neural network performs a sigma-pi function to collect inputs. To obtain this function, the templates used in the algorithm are transformed to equivalent Boolean expressions. After the neural network converges, a perfectly 8-connected skeleton is derived. Good experimental results show the feasibility of the proposed approach.
3 citations