Journal ArticleDOI
A Comparative Study of Texture Measures for Terrain Classification
Joan S. Weszka,Charles R. Dyer,Azriel Rosenfeld +2 more
- Vol. 6, Iss: 4, pp 269-285
Reads0
Chats0
TLDR
In this paper, three standard approaches to automatic texture classification make use of features based on the Fourier power spectrum, on second-order gray level statistics, and on first-order statistics of gray level differences, respectively.Abstract:
Three standard approaches to automatic texture classification make use of features based on the Fourier power spectrum, on second-order gray level statistics, and on first-order statistics of gray level differences, respectively. Feature sets of these types, all designed analogously, were used to classify two sets of terrain samples. It was found that the Fourier features generally performed more poorly, while the other feature sets all performned comparably.read more
Citations
More filters
Book ChapterDOI
Digital Image Processing and Recognition
TL;DR: This paper reviews some of the recent developments in image recognition techniques, including data structures for image analysis; image matching; segmentation; texture analysis; and shape description.
Proceedings ArticleDOI
Detection of Urban Zones in Satellite Images using Visual Words
Lior Weizman,Jacob Goldberger +1 more
TL;DR: This paper addresses the problem of urban areas extraction by using a visual representation concept known as "Bag of Words", originally developed for text retrieval approaches, and is robust to changes in atmospheric conditions during acquisition time.
Proceedings ArticleDOI
Shift and rotation invariant texture recognition with neural nets
M.A. Mayorga,L.C. Ludeman +1 more
TL;DR: Shift and rotational invariant approaches based on neural nets have been proposed in the literature for the purpose of a training for classification of patterns independently of their position and orientation and this approach leads to a considerable time saving in the learning phase of the net.
Book ChapterDOI
SVM and haralick features for classification of high resolution satellite images from urban areas
Aissam Bekkari,Soufiane Idbraim,Azeddine Elhassouny,Driss Mammass,Mostafa El Yassa,Danielle Ducrot +5 more
TL;DR: A methodology allowing to combine these two informations using a combination of multi-spectral features and Haralick texture features as data source with composite kernel is proposed which allows a significant improvement of the classification performances when compared with the two sets of attributes used separately.
Book ChapterDOI
Image Segmentation and Pattern Recognition: A Novel Concept, the Histogram of Connected Elements
TL;DR: This chapter aims to model the image segmentation process as a pattern recognition problem, which implies that any method or technique from the pattern recognition field can, in principle, be applied to solve the segmentation problem in any computer-based vision system or application.
References
More filters
Journal ArticleDOI
Textural Features for Image Classification
TL;DR: These results indicate that the easily computable textural features based on gray-tone spatial dependancies probably have a general applicability for a wide variety of image-classification applications.
Journal ArticleDOI
Texture analysis using gray level run lengths
TL;DR: In this paper, a set of texture features based on gray level run lengths is described, and good classification results are obtained with these features on a sets of samples representing nine terrain types.
Proceedings Article
Computer description of textured surfaces
TL;DR: This work deals with computer analysis of textured surfaces with descriptions of textures formalized from natural language descriptions obtained from the directional and non-directional components of the Fourier transform power spectrum.
Spectral and textural processing of ERTS imagery
R. M. Haralick,R. J. Bosley +1 more
TL;DR: In this article, a procedure is developed to simultaneously extract textural features from all bands of ERTS multispectral scanner imagery for automatic analysis, and an ellipsoidally symmetric functional form is assumed for the co-occurrence distribution of multiimage greytone N-tuple differences.