Combined Statistical and Structural Approach for Unsupervised Texture Classification
نویسنده
چکیده
In this paper a combined statistical and structural approach has been employed for texture representation. A set of Texture Primitives has been suggested. These primitives are basically tested for the presence of texture by conducting a suitable statistical test called Nair’s test. The set of universal primitives are labeled as local descriptor and the frequency of occurrences of these primitives is used as the global descriptor, namely Texture Primitive Spectrum for a given texture image. Since the occurrence of primitives and their placement rules uniquely define a texture image, the primitive spectrum is also unique, for a texture image. These spectrums are shown to be effectively used for un supervised texture classification for Brodatz and Vistex data bases of texture images. An average correct classification of 96% has been obtained. Key-words TexturePrimitives – Universal set of Primitives – global descriptor Texture Primitive Spectrum – Unsupervised texture classification.
منابع مشابه
Visualizing 3 - D Texture : a Three Dimensional Structural
The evaluation of texture in digital remote sensing, is traditionally accomplished by applying statistical, rather than structural models. This approach is partially due to the difficulty in determining structural placement rules for natural scenes, and to early psychophysical texture theories that demonstrated humans were sensitive to second-order statistics. A literature review indicates that...
متن کاملSpectral-spatial classification of hyperspectral images by combining hierarchical and marker-based Minimum Spanning Forest algorithms
Many researches have demonstrated that the spatial information can play an important role in the classification of hyperspectral imagery. This study proposes a modified spectral–spatial classification approach for improving the spectral–spatial classification of hyperspectral images. In the proposed method ten spatial/texture features, using mean, standard deviation, contrast, homogeneity, corr...
متن کاملUnsupervised texture classification : Automatically discover and classify texture patterns q
In this paper, we present a novel approach to classify texture collections. This approach does not require experts to provide annotated training set. Given the image collection, we extract a set of invariant descriptors from each image. The descriptors of all images are vector-quantized to form ‘keypoints’. Then we represent the texture images by ‘bag-of-keypoints’ vectors. By analogy text clas...
متن کاملTexture Classification Using Combined Statistical Approach
The statistical approaches such as texture spectrum and local binary pattern methods have been discussed in this paper. The features are extracted by the computation of LBP and Texture Spectrum histogram. A combined approach of LBP with texture spectrum is also proposed further. Experiments of texture feature extraction, classification of textures and similarity-based image-to-image matching ar...
متن کاملTexture Classification by Combining Local Binary Pattern Features and a Self-Organizing Map
This paper deals with the combined use of Local Binary Pattern (LBP) features and a Self-Organizing Map (SOM) in texture classification. With this approach, the unsupervised learning and visualization capabilities of a SOM are utilized with highly efficient histogram-based texture features. In addition to the Euclidean distance normally used with a SOM, an information theoretic log-likelihood (...
متن کامل