Brief review of invariant texture analysis methods

Abstract

This paper considers invariant texture analysis. Texture analysis approaches whose performances are not a ected by translation, rotation, a ne, and perspective transform are addressed. Existing invariant texture analysis algorithms are carefully studied and classiÿed into three categories: statistical methods, model based methods, and structural methods. The importance of invariant texture analysis is presented ÿrst. Each approach is reviewed according to its classiÿcation, and its merits and drawbacks are outlined. The focus of possible future work is also suggested. ?

A Weakly Supervised Approach for Semantic Image Indexing and Retrieval

“…Numerical features are associated with visual concepts and define how visual concepts are computed on image data. Examples of numerical features associated with visual concept are : color coherence vectors [7] for visual concept Hue; co-occurence matrices [8] for visual concept P attern; Sift features [9] and Mpeg-7 shape features [10] for visual concept Geometry.…”

Invariant texture segmentation via circular Gabor filters

“…In recent years, invariant texture analysis has been paid more and more attention due to its increasing importance. A great deal of wok has been done on this topic [2][3] [14] [15]. However most of the existing methods focus on invariant texture classification.…”

Color texture analysis based on fractal descriptors

“…Many of these approaches represent the local behavior of the texture via statistical [1], structural [2] or spectral [3][4][5] properties of the image. Good surveys can be found in [5][6][7][8][9][10][11]. The conjecture presented in [12], where second-order probability distributions [6,13] are enough for human discrimination of two texture patterns, has motivated the use of statistical approaches.…”