On the performance of Some Edge Detectors for Gray Scale Images

Abstract

Image processing represents an important challenge in different fields. Mathematical Morphology uses concepts from set theory, geometry, algebra and topology to analyze the geometrical structure of an image. Using them, powerful tools as dilation and erosion operators are introduced to solve problems related to edge detection or image segmentation among others. In addition, fuzzy relations are also a useful tool for image processing, primarily those methods based on interval-valued fuzzy relations. Actually, they can be understood as a gradient from a morphological point of view, although they are not dilations neither erosions. In this work a comparison between different approaches is performed,checking some configurations and evaluating in terms of least squares fitting the configuration with the best performance.