This paper presents a theoretically very simple yet efficient approach for gray scale and rotation invariant texture classification based on local binary patterns and nonparametric discrimination of sample and prototype distributions. The proposed approach is very robust in terms of gray scale variations, since the operators are by definition invariant against any monotonic transformation of th...

A fully parallel iterative thinning algorithm called MB2 is presented. It favourably competes with the best known algorithms regarding homotopy, mediality; thickness, rotation invariance and noise immunity, while featuring a speed improvement by a factor two or more owing to a smaller number of operations to perform. MB2 is grounded on a simple physics-based thinning principle that conveys both...

We first describe a simple method for diagonalizing the isotropic exchange Hamil, tonian of a cluster of N spins in the most general case where all the exchange constants are different. The technique, based on the rotation invariance of the system, leads to a considerable reduction of the total matrix. Simple expressions of the magnetization and susceptibility are provided and an example of the...

We generalize the study of symbolic dynamical systems of nite type and ZZ 2 action, and the associated use of symbolic substitution dynamical systems, to dynamical systems with IR 2 action. The new systems are associated with tilings of the plane. We generalize the classical technique of the matrix of a substitution to include the geometrical information needed to study tilings, and we utilize ...

Due to the variety of shapes and sizes that present both living organisms and static objects, the necessity to look for automated systems of identification, both in industry and scientific research has arisen. During the 1960s, the scientific community in the field of optics has used the Fourier transform and other types of mathematical transformations for pattern recognition, taking advantage ...

• Rotational invariance, SO(3). Here we can rotate through an arbitrary angle specified by a continuous parameter θ, restricted to some finite range, say [0, 2π). There are a infinite continuum of possible values of θ, even though its range is limited. There are also two continuous parameters necessary to specify the direction about which this rotation is to take place. This is a three-paramete...

A Gabor filter based feature extraction scheme for texture classification is proposed. By using a novel set of circularly symmetric filters, rotation invariance is achieved. The scheme offers a high classification performance on textures at any orientation using both fewer features and a smaller area of analysis than most existing schemes. The performance of the scheme on noisy images is also i...

We present a method to recognize three-dimensional objects from phase-shift digital holograms. The holograms are used to reconstruct various views of the objects. These views are combined to create nonlinear composite filters in order to achieve distortion invariance. We present experiments to illustrate the recognition of a 3D object in the presence of outof-plane rotation and longitudinal shi...

The present paper is devoted to the introduction of a novel technique to select keypoints from digital images and build representative and distinctive descriptors. The algorithm performs a multiresolution image analysis in the Laguerre Gauss transformed domain and collects in a local descriptor the transformed coefficients at multiple scales, of the keypoint’s representative pattern. The rotati...

The sub-block-based image retrieval method utilizes global and local image features to retrieve trademarks through separating individual an images into some blocks. This paper proposes a sub-block-based trademark image retrieval algorithm under the polar coordinate system. Experiment results show that our algorithm can keep excellent invariance in image translation, scaling and rotation and the...