Texture anisotropy is a very important cue in object recognition. Its importance from the physiological point of view is underlined by the fact that the mammalian vision systems include orientation-selective mechanisms in the visual cortex [12, 13], and from the psychophysical point of view by the fact that orientation is one of the cues used by humans in the perceptual grouping of patterns [14,24]. Gorkani and Picard  showed that a simple measure, such as the "dominant perceived orientation," may suffice for the quick coarse classification of certain kinds of image scenes. For example, it is known that pigeons classify cities and countries in a similar way . Sato and Cipolla used the moments of local texture to perform image registration and estimate surface orientation exploiting the anisotropy induced by projection .
The importance of anisotropy in materials has also been widely recognized, and so it has been studied by several methods in many other disciplines. For example, anisotropy can be used as a diagnostic tool for the identification of faults  and the behavior of the formation process of the imaged structure [2, 15, 17]. In mechanics, anisotropy has been estimated with the help of angular scattering of coherent light across rough surfaces , the calculation of the maximum to minimum ratio of radii of the covariance function computed for several azimuthal directions , etc. Anisotropy characterization at many scales has also been used for the analysis of engineering surfaces  and tumors in liver ultrasound images .
Was this article helpful?