Detecting image primitives using feature pyramids

https://doi.org/10.1016/S0020-0255(97)10021-4Get rights and content

Abstract

Primitive image parts of arbitrary sizes can be detected and extracted efficiently using local operations in a multiscale (“pyramid”) representation of the image. This paper describes several new multiscale methods of detecting blob-like image parts, and also presents some initial results on the detection of ribbon-like image parts.

References (28)

  • T.N. Cornsweet

    Visual Perception

    (1970)
  • A.D. Gross et al.

    Multiresolution object detection and delineation

    Computer Vision, Graphics, and Image Processing

    (1987)
  • G. Hartmann

    Recognition of hierarchically encoded images by technical and biological systems

    Biological Cybernetics

    (1987)
  • Cited by (9)

    • Some new topological properties of the triangular pyramid networks

      2013, Information Sciences
      Citation Excerpt :

      A processor interconnection network or a communications network can be modeled by a graph G, in which every vertex corresponds to a processor or a switching element, and every edge corresponds to a communication link. A pyramid network (abbreviated to pyramid) is one of the important network topologies, as it has been used as both a hardware architecture and a software structure for parallel and network computing, image processing, and computer vision [3,11,21,22,31]. A new pyramidal network, the triangular pyramid (abbreviated to tripy), was proposed by Razavi and Sarbazi-Azad in [30].

    • Using resolution pyramids for watershed image segmentation

      2007, Image and Vision Computing
      Citation Excerpt :

      G1 is interpreted as a 3D landscape, where the grey-level g of the pixel in position (x, y), is used as the third coordinate in the landscape. Pyramids are well known multi-resolution representation systems; they provide from coarse to fine representations of a discrete image [7], and have been employed for a number of applications, such as line-drawing analysis or object contour extraction [8–13] and segmentation [14–18]. The general regular pyramid construction strategy is based on the use of a uniform subdivision rule that associates to fixed size regions in the image at a given resolution, single pixels in the image at immediately lower resolution.

    • Image analysis and computer vision: 1998

      1999, Computer Vision and Image Understanding
    • Acquired brain injury cognitive dysfunctional profile based on neuropsychological knowledge and medical imaging studies

      2014, 2014 IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2014
    View all citing articles on Scopus
    1

    Present address: Federal Highway Administration, McLean, VA 22101, USA.

    View full text