STRUCTURAL, DATA DRIVEN EDGE DETECTION

摘要

The paper presents a general feature detection method. The aim of the research is to create a method in which the operating software code and the data are separated. The method is based on a new object, the disk definition. The feature detection is reduced to a picture element definition and an optimization problem. The algorithm can be used as a building block in a general graph based object recognition system. Through two simple examples we demonstrate how to solve practical problems with the presented method; a line detection and gray level edge detection problem is solved.