Skip to Main content Skip to Navigation
Theses

Some Geometric Methods for the Analysis of Images and Textures

Abstract : This thesis focuses on the studies of the extraction and characterization of local image structures, in the context of images and texture analysis. Relying on the level lines of images or on the somehow dual and less structured notion of gradient orientation, the contributions of the thesis concentrate on following three themes: The first part of this thesis presents a new method for texture analysis that in spirit is similar to morphological granulometries, while allowing a high degree of geometrical and radiometric invariances. Also using the topographic map representation, the second part of this thesis develops a general approach for the abstraction of images, the aim of which is to automatically generate abstract images from realistic photographs. The subject of the last part of this thesis is the detection of junctions in natural images. The approach relies on the local directions of level lines through the orientation of image gradient. We introduce a generic junction analysis scheme. The first asset of the proposed procedure is an automatic criterion for the detection of junctions, permitting to deal with textured parts in which no detection is expected. Second, the method yields a characterization of L-, Y- and X- junctions, including a precise computation of their type, localization and scale. Contrary to classical approaches, scale characterization does not rely on the linear scale-space, therefore enabling geometric accuracy.
Complete list of metadatas

Cited literature [226 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00682590
Contributor : Gui-Song Xia <>
Submitted on : Monday, March 26, 2012 - 1:43:04 PM
Last modification on : Friday, July 31, 2020 - 10:44:07 AM

Identifiers

  • HAL Id : pastel-00682590, version 1

Collections

Citation

Gui-Song Xia. Some Geometric Methods for the Analysis of Images and Textures. Computer Vision and Pattern Recognition [cs.CV]. Télécom ParisTech, 2011. English. ⟨pastel-00682590⟩

Share

Metrics

Record views

765

Files downloads

935