Author_Institution :
Dept. of Electr. Eng., Univ. of New Orleans, LA, USA
Abstract :
This paper introduces a technique for synthesizing natural textures, with emphasis on quasiperiodic and structural textures. Textures are assumed to be composed of three components, namely illumination, structure, and stochastic. The contribution of this work is that, in contrast to previous techniques, it proposes a joint approach for handling the texture´s global illumination, irregular structure, and stochastic component which may be correlated to the other two components. Furthermore, the proposed technique does not produce verbatim copies in the synthesized texture. More specifically, a top-down approach is used for extraction of texture elements (textons) in which, in contrast to previous texton-based approaches, no assumptions regarding perfect periodicity are made. The structure itself can be modeled as a stochastic process. Consequently, textons are allowed to have irregular and nonidentical shapes. In the synthesis stage, a new nonregular textural structure is designed from the original one that defines the place holders for textons. We call such place holders empty textons (e-textons). The e-textons are filled in by a representative texton. Since e-textons do not have identical shapes, a texton shape-matching procedure is required. After adding the illumination to the structural component, a strictly localized version of a block sampling technique is applied to add the stochastic component. The block sampling technique combined with the addition of the illumination component provides a significant improvement in the appearance of synthesized textures. Results show that the proposed method is successful in synthesizing structural textures visually indistinguishable to the original. Moreover, the method is successful in synthesizing a variety of stochastic textures.
Keywords :
image matching; image sampling; image texture; stochastic processes; block sampling technique; periodicity detection; quasiperiodic texture; shape-matching procedure; stochastic process; texton-based approaches; texture elements extraction; texture synthesis; top-down approach; Application software; Computer applications; Computer vision; Image segmentation; Lighting; Predictive models; Sampling methods; Shape; Spatial resolution; Stochastic processes; Periodicity detection; textons; texture synthesis; Algorithms; Image Enhancement; Image Interpretation, Computer-Assisted; Imaging, Three-Dimensional; Information Storage and Retrieval;