Edge Detection based on Kernel Density Estimation
This addresses edge detection for computer vision tasks, but it appears incremental as it builds on existing kernel-based approaches.
The paper tackles edge detection in images by proposing a method based on kernel density estimation, where pixels with minimum density are labeled as edges, and experimental evaluations show it is a competitive algorithm.
Edges of an image are considered a crucial type of information. These can be extracted by applying edge detectors with different methodology. Edge detection is a vital step in computer vision tasks, because it is an essential issue for pattern recognition and visual interpretation. In this paper, we propose a new method for edge detection in images, based on the estimation by kernel of the probability density function. In our algorithm, pixels in the image with minimum value of density function are labeled as edges. The boundary between two homogeneous regions is defined in two domains: the spatial/lattice domain and the range/color domain. Extensive experimental evaluations proved that our edge detection method is significantly a competitive algorithm.