Stefano Casadei

Learn More
The problem of edge detection is viewed as a hierarchy of detection problems where the geometric objects to be detected (e.g. edge points, curves, regions) have increasing complexity and spatial extent. An early stage of the proposed hierarchy consists in detecting the regular portions of the visible edges. The input to this stage is given by a graph whose(More)
We address the problem of scale selection in texture analysis. Two diierent scale parameters, feature scale and statistical scale, are de-ned. Statistical scale is the size of the regions used to compute averages. We deene the class of homogeneous random functions as a model of texture. A dishomogeneity function is deened and we prove that it has useful(More)
EEcient edge detection algorithms such as Canny's fail near curve singularities. Moreover, the standard linking algorithms used on top of these detectors often fail because of instabilities in the tracking process (due to multiple responses to the same edge and interference of nearby edges). We propose a hierarchical approach to edge detection based on a(More)
Conventional edge linking methods perform poorly when multiple responses to the same edge, bifurcations and nearby edges are present. We propose a scheme for curve inference where divergent bifur-cations are initially suppressed so that the smooth parts of the curves can be computed more reliably. Recovery of curve singularities and gaps is deferred to a(More)
Perceptual organization provides an intermediate representation of data by means of object-and goal-independent information. The lack of complete information makes perceptual organization an in-trinsically ambiguous process which invalidates the uniqueness assumption and requires instead the generation of multiple solutions. This raises the issue of(More)
—A large portion of image contours is characterized by local properties such as sharp variations of the image intensity across the contour. The integration of local image descriptors estimated by using these local properties into curvilinear descriptors is a difficult problem from a theoretical viewpoint because of the combinatorially large number of(More)
  • 1