Orientation- and scale-invariant recognition of textures in multi-object scenes

Abstract

This contribution describes a novel approach to orientation and scale-invariant detection of textured objects in images. It performs both, a segmentation of multi-object scenes and the identification of rotation angles and scale rates of textures in an image by applying comparison with reference texture features stored in a database. The main novelty of the proposed method is the transform of rotation and dilation into shifts in the feature space by employing a polar-log Gabor filter bank. Texture segmentation and identification of the rotation angles and scale rates have been carried out using symmetric phase only matched filters. The simulation results illustrated in this paper highlight the performance of the presented method in an exemplary manner

Image labelling in real conditions

“…When the aim is the segmentation and interpretation of the objects in a real scene, the task can be made easier if the configuration at surface texture level is known (Won, 2000;Bhalerao and Wilson, 2000;Campbell et al, 1997;Shanahan et al, 1999). Many techniques highlight the classification capacity of the characterizers extracted from images (Haralick, 1979;Tamura et al, 1978) searching for properties that are invariable or tolerant with the variation of optic parameters (Cohen et al, 1991;Leow and Lai, 2000;Sim et al, 2000;Teuner et al, 1997).…”

Analysis of orientation and scale in smoothly varying textures

“…For example, [3] has looked at using wavelets and [4] utilizes Gabor filters. The work in [14] presents an approach that uses a set of Gabor filters, observing that changes in scale and orientation are manifested as shifts in their feature space. Montoya-Zegarra et al [10] proposed a method using steerable pyramids [6] where they treat each filter output independently.…”

“…For example, [10] and [14] represent each texture with a constant measure within a region. In contrast to the method presented here, these methods either do not consider variations in scale and orientation or treat them as nuisance parameters.…”