This paper deals with video segmentation based on motion and spatial information. Classically;the nucleus of the motion term is the motion compensation error (NICE) between two consecutive frames. Defining a motion-ba...
详细信息
ISBN:
(纸本)9783540728221
This paper deals with video segmentation based on motion and spatial information. Classically;the nucleus of the motion term is the motion compensation error (NICE) between two consecutive frames. Defining a motion-based energy as the integral of a, function of the NICE over the object domain implicitly results in making an assumption on the MCE distribution: Gaussian for the square function, Laplacian for the absolute value, or other parametric distributions for functions used in robust estimation. However, these assumptions are generally false. Instead, it is proposed to integrate a function of (an estimation of) the NICE distribution. The function is taken such that the integral is the Ahmad-Lin entropy of the MCE, the purpose being to be more robust to outliers. Since a motion-only approach can fail in homogeneous areas, the proposed energy is the joint entropy of the NICE and the object color. It is minimized using active contours.
People often recognize 3D objects by their boundary shape. Designing an algorithm for such a task is interesting and useful for retrieving objects from a shape database. In this paper, we present a fast 2-stage algori...
详细信息
ISBN:
(纸本)9783540728221
People often recognize 3D objects by their boundary shape. Designing an algorithm for such a task is interesting and useful for retrieving objects from a shape database. In this paper, we present a fast 2-stage algorithm for recognizing 3D objects using a new feature space, built from curvature scalespace images, as a shape representation that is scale, translation, rotation and reflection invariant. As well, the new shape representation removes the inherent ambiguity of the zero position of arc length for a scalespace image. The 2-stage matching algorithm, conducted in the eigenspaces of the feature space, is analogous to the way people recognize an object: first identifying the type of object, and then determining the actual object. We test the new algorithm on a 3D database comprising 209 colour objects in 2926 different view poses, and achieve a 97% recognition rate for the object type and 95% for the object pose.
In the standard scalespace approach one obtains a scalespace representation u : R(d) x R(+) -> R of an image f epsilon L(2) (R(d)) by means of an evolution equation on the additive group (R(d), +). However, it is...
详细信息
ISBN:
(纸本)9783540728221
In the standard scalespace approach one obtains a scalespace representation u : R(d) x R(+) -> R of an image f epsilon L(2) (R(d)) by means of an evolution equation on the additive group (R(d), +). However, it is common to apply a wavelet transform (constructed via a representation U of a Lie-group G and admissible wavelet psi) to an image which provides a detailed overview of the group structure in an image. The result of such a wavelet transform provides a function g -> (U(g)psi, f)(L2(R)(2)()) on a group G (rather than (R(d), +)), which we call a score. Since the wavelet transform is unitary we have stable reconstruction by its adjoint. This allows us to link operators on images to operators on scores in a robust way. To ensure U-invariance of the corresponding operator on the image the operator on the wavelet transform must be left-invariant. Therefore we focus on left-invariant evolution equations (and their resolvents) on the Lie-group G generated by a quadratic form Q on left invariant vector fields. These evolution equations correspond to stochastic processes on G and their solution is given by a group convolution with the corresponding Green's function, for which we present an explicit derivation in two particular image analysis applications. In this article we describe a general approach how the concept of scalespace can be extended by replacing the additive group R(d) by a Lie-group with more structure.
In clinical applications where structural asymmetries between homologous shapes have been correlated with pathology, the questions of definition and quantification of 'asymmetry' arise naturally. When not only...
详细信息
ISBN:
(纸本)9783540728221
In clinical applications where structural asymmetries between homologous shapes have been correlated with pathology, the questions of definition and quantification of 'asymmetry' arise naturally. When not only the degree but the position of deformity is thought relevant;asymmetry localization must also be addressed. Asymmetries between paired shapes have already been formulated in terms of (non-rigid) diffeomorphisms between the shapes. For the infinity of such maps possible for a given pair, we define optimality as the minimization of deviation from isometry under the constraint of piecewise deformation homogeneity. We propose a novel variational formulation for segmenting asymmetric regions from surface pairs based on the minimization of a functional of both the deformation map and the segmentation boundary;which defines the regions within which the homogeneity constraint is to be enforced. The functional minimization is achieved via a quasi-simultaneous evolution of the map and the segmenting curve, conducted on and between two-dimensional surface parametric domains. We present examples using both synthetic data and pairs of left and right hippocampal structures;and demonstrate the relevance of the extracted features through a clinical epilepsy classification analysis.
Nowadays image acquisition in materials science allows the resolution of grains at atomic scale. Grains are material regions with different lattice orientation which are typically not in equilibrium. At the same time,...
详细信息
ISBN:
(纸本)9783540728221
Nowadays image acquisition in materials science allows the resolution of grains at atomic scale. Grains are material regions with different lattice orientation which are typically not in equilibrium. At the same time, new microscopic simulation tools allow to study the dynamics of such grain structures. Single atoms are resolved in the experimental and in the simulation results. A qualitative study of experimental images and simulation results and the comparison of simulation and experiment requires the robust and reliable extraction of mesoscopic properties from these microscopic data sets. Based on a Mumford-Shah type functional, grain boundaries are described as free discontinuity sets at which the orientation parameter for the lattice jumps. The lattice structure itself is encoded in a suitable integrand depending on the local lattice orientation. In addition the approach incorporates solid-liquid interfaces. The resulting Mumford-Shah functional is approximated with a level set active contour model following the approach by Chan and Vese. The implementation is based on a finite element discretization in space and a,step size controlled gradient descent algorithm.
Partial Differential equations (PDE), wavelets-based methods and neighborhood filters were proposed as locally adaptive machines for noise removal. Recently, Buades, Coll and Morel proposed the Non-Local (NL-) means f...
详细信息
ISBN:
(纸本)9783540728221
Partial Differential equations (PDE), wavelets-based methods and neighborhood filters were proposed as locally adaptive machines for noise removal. Recently, Buades, Coll and Morel proposed the Non-Local (NL-) means filter for image denoising. This method replaces a noisy pixel by the weighted average of other image pixels with weights reflecting the similarity between local neighborhoods of the pixel being processed and the other pixels. The NL-means filter was proposed as an intuitive neighborhood filter but theoretical connections to diffusion and non-parametric estimation approaches are also given by the authors. In this paper we propose another bridge, and show that the NL-means filter also emerges from the Bayesian approach with new arguments. Based on this observation, we show how the performance of this filter can be significantly improved by introducing adaptive local dictionaries and a new statistical distance measure to compare patches. The new Bayesian NL-means filter is better parametrized and the amount of smoothing is directly determined by the noise variance (estimated from image data) given the patch size. Experimental results are given for real images with artificial Gaussian noise added, and for images with real image-dependent noise.
In this work a marker-controlled and regularized watershed segmentation is proposed. Only a few previous studies address the task of regularizing the obtained watershed lines from the traditional marker-controlled wat...
详细信息
ISBN:
(纸本)9783540728221
In this work a marker-controlled and regularized watershed segmentation is proposed. Only a few previous studies address the task of regularizing the obtained watershed lines from the traditional marker-controlled watershed segmentation. In the present formulation, the topographical distance function is applied in a level set formulation to perform the segmentation, and the regularization is easily accomplished by regularizing the level set functions. Based on the well-known Four-Color theorem, a mathematical model is developed for the proposed ideas. With this model, it is possible to segment any 2D image with arbitrary number of phases with as few as one or two level set functions. The algorithm has been tested on real 2D fluorescence microscopy images displaying rat cancer cells, and the algorithm has also been compared to a standard watershed segmentation as it is implemented in MATLAB. For a fixed set of markers and a fixed set of challenging images, the comparison of these two methods shows that the present level set formulation performs better than a standard watershed segmentation.
We present a new flexible wavefront propagation algorithm for the boundary value problem for sub-Riemannian (SR) geodesics in the roto-translation group SE(2) = 2 S1 with a metric tensor depending on a smooth external...
详细信息
暂无评论