G06V 10/54

Definition

Diese Klassifikationsstelle umfasst:

(Für diese Definition ist die deutsche Übersetzung noch nicht abgeschlossen)

Texture feature extraction for image or video recognition or understanding, either by identifying the boundaries of texture regions, or by analysing the content of the regions themselves.

Notes – technical background

These notes provide more information about the technical subject matter that is classified in this place:

Examples of algorithms used for feature extraction include:

Statistical approaches which characterise the texture by local statistical measures such as “edgeness” (local variation of the image gradient), co-occurrence matrices and Haralick features, Laws texture energy, local histogram-based measures, autocorrelation, power spectrum, etc.;
Structural approaches based on primitives, morphological operations or representations derived from them, or graph-based methods in which image quantities (e.g. pixels or local patches) are represented as graph nodes and are clustered together using graph-based clustering algorithms (e.g. graph-cuts) to identify texture regions;
Model-based approaches such as auto-regressive models, fractal models, random fields, texton model;
Transform methods such as Fourier (spectral) analysis, Gabor filters, wavelets, curvelet transform.

Examples

Bildreferenz:G06V0010540000_0

Texture feature extraction allows identification of an animal (zebra) in natural images.

Querverweise

Informative Querverweise

Global feature extraction by analysis of the whole pattern	G06V 10/42
Local feature extraction by analysis of the parts of the pattern, e.g. by detecting edges, contours, loops, corners, intersections; Connectivity analysis, e.g. connected component analysis	G06V 10/44
Descriptors for shape, contour or point-related descriptors, e.g. SIFT	G06V 10/46
Colour feature extraction	G06V 10/56
Feature extraction related to illumination properties	G06V 10/60
Pattern recognition or image understanding, using clustering	G06V 10/762
Analysis of texture in general	G06T 7/40

Glossar

GLCH	grey-level co-occurrence histogram (synonym of GLCM)
GLCM	grey-level co-occurrence matrix (Haralick invariant texture features)
Texton	basic component of an image that may be recognised visually before the entire image is recognised, and that repeats itself to generate a texture region.