Results 1  10
of
21
Multiresolution histograms an their use for recognition.
 IEEE Transactions on Pattern Analysis and Machine Intelligence,
, 2004
"... ..."
(Show Context)
On the Efficient Evaluation of Probabilistic Similarity Functions for Image Retrieval
 IEEE Trans. Inf. Theory
, 2004
"... Probabilistic approaches are a promising solution to the image retrieval problem that, when compared to standard retrieval methods, can lead to a significant gain in retrieval accuracy. However, this occurs at the cost of a significant increase in computational complexity. In fact, closedform solut ..."
Abstract

Cited by 36 (1 self)
 Add to MetaCart
Probabilistic approaches are a promising solution to the image retrieval problem that, when compared to standard retrieval methods, can lead to a significant gain in retrieval accuracy. However, this occurs at the cost of a significant increase in computational complexity. In fact, closedform solutions for probabilistic retrieval are currently available only for simple probabilistic models such as the Gaussian or the histogram. We analyze the case of mixture densities and exploit the asymptotic equivalence between likelihood and Kullback–Leibler (KL) divergence to derive solutions for these models. In particular, 1) we show that the divergence can be computed exactly for vector quantizers (VQs) and 2) has an approximate solution for Gauss mixtures (GMs) that, in highdimensional feature spaces, introduces no significant degradation of the resulting similarity judgments. In both cases, the new solutions have closedform and computational complexity equivalent to that of standard retrieval approaches.
Detection, Synthesis and Compression in Mammographic Image Analysis with a Hierarchical Image Probability Model
 In L. Staib (editor), IEEE Workshop on Mathematical Methods in Biomedical Image Analysis. 2001
"... All intext references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately. ..."
Abstract

Cited by 20 (4 self)
 Add to MetaCart
(Show Context)
All intext references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
Towards Efficient Texture Classification and Abnormality Detection
, 2004
"... One of the fundamental issues in image processing and machine vision is texture, specifically texture feature extraction, classification and abnormality detection. This thesis is concerned with the analysis and classification of natural and random textures, where the building elements and the struct ..."
Abstract

Cited by 13 (2 self)
 Add to MetaCart
One of the fundamental issues in image processing and machine vision is texture, specifically texture feature extraction, classification and abnormality detection. This thesis is concerned with the analysis and classification of natural and random textures, where the building elements and the structure of texture are not clearly determinable, hence statistical and signal processing approaches are more appropriate. We investigate the advantages of multiscale/multidirectional signal processing methods, higher order statisticsbased schemes, and computationally low cost texture analysis algorithms. Consequently these advantages are combined to form novel algorithms.
Hierarchical Image Probability (HIP) Models
 in Adv. Neural Information Processing Systems
, 2000
"... We formulate a model for probability distributions on image spaces. We show that any distribution of images can be factored exactly into conditional distributions of feature vectors at one resolution (pyramid level) conditioned on the image information at lower resolutions. We would like to factor t ..."
Abstract

Cited by 8 (2 self)
 Add to MetaCart
(Show Context)
We formulate a model for probability distributions on image spaces. We show that any distribution of images can be factored exactly into conditional distributions of feature vectors at one resolution (pyramid level) conditioned on the image information at lower resolutions. We would like to factor this over positions in the pyramid levels to make it tractable, but such factoring may miss longrange dependencies. To capture longrange dependencies, we introduce hidden class labels at each pixel in the pyramid. The result is a hierarchical mixture of conditional probabilities, similar to a hidden Markov model on a tree. The model parameters can be found with maximum likelihood estimation using the EM algorithm. We have obtained encouraging preliminary results on the problems of detecting various objects in SAR images and target recognition in optical aerial images. 1. INTRODUCTION Many approaches to object recognition in images estimate Pr(C j I), the probability that an object of clas...
A MultiScale Probabilistic Network Model for Detection, Synthesis and Compression in Mammographic Image Analysis
, 2003
"... We develop a probabilistic network model over image spaces and demonstrate its broad utility in mammographic image analysis, particularly with respect to computeraided diagnosis. The model employs a multiscale pyramid decomposition to factor images across scale and a network of treestructured hid ..."
Abstract

Cited by 6 (1 self)
 Add to MetaCart
We develop a probabilistic network model over image spaces and demonstrate its broad utility in mammographic image analysis, particularly with respect to computeraided diagnosis. The model employs a multiscale pyramid decomposition to factor images across scale and a network of treestructured hidden variables to capture longrange spatial dependencies. This factoring makes the computation of the density functions local and tractable. The result is a hierarchical mixture of conditional probabilities, similar to a hidden Markov model on a tree. The model parameters are found with maximum likelihood estimation using the expectationmaximization algorithm. The utility of the model is demonstrated for three applications: (1) detection of mammographic masses for computeraided diagnosis; (2) qualitative assessment of model structure through mammographic synthesis; and (3) compression of mammographic regions of interest.
Morphological color size distributions for Image classification and retrieval
 Proceeding of ACIVS (Advanced Concept for Intelligent Vision Systems
, 2002
"... Current contentbased image retrieval techniques can typically perform efficient and effective searches on heterogeneous image databases. This contribution deals with an approach based on the integration of color and texture description which is applied to a very homogeneous database: a blood ima ..."
Abstract

Cited by 6 (2 self)
 Add to MetaCart
(Show Context)
Current contentbased image retrieval techniques can typically perform efficient and effective searches on heterogeneous image databases. This contribution deals with an approach based on the integration of color and texture description which is applied to a very homogeneous database: a blood image bank. The content of images is very similar and therefore it becomes imperative to use very precise descriptors: the color is described by classical color distributions (histograms) and for the texture, we introduce the morphological color size distributions. The similarity is measured by computing distance metrics between histograms. In order to increase the accuracy of retrieval, the results of colorbased and texturebased retrieval are integrated by combining the associated dissimilarity values. The effects of different integration methods on classification performance are explained by means of experimental tests in a database of 123 cell images (leukocyte color images). After learning processing, where different feature selection and classifier definition alternatives are tested, a definitive integrated approach is proposed (precision ). 1.
Nonparametric Estimation of Aspect Dependence for ATR
 in Algorithms for Synthetic Aperture Radar Imagery VI
, 1999
"... In conventional SAR image formation, idealizations are made about the underlying scattering phenomena in the target field. In particular, the reflected signal is modeled as a pure delay and scaling of the transmitted signal where the delay is determined by the distance to the scatterer. Inherent in ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
(Show Context)
In conventional SAR image formation, idealizations are made about the underlying scattering phenomena in the target field. In particular, the reflected signal is modeled as a pure delay and scaling of the transmitted signal where the delay is determined by the distance to the scatterer. Inherent in this assumption is that the scatterers are isotropic, i.e. their reflectivity appears the same from all orientations, and frequency independent, i.e. the magnitude and phase of the reflectivity are constant with respect to the frequency of the transmitted signal. Frequently, these assumptions are relatively poor resulting in an image which is highly variable with respect to imaging aspect. This variability often poses a difficulty for subsequent processing such as ATR. However, this need not be the case if the nonideal scattering is taken into account. In fact, we believe that if utilized properly, these nonideal characteristics may actually be used to aid in the processing as they convey di...
Varying Complexity in TreeStructured Image Distribution Models
 IEEE TRANSACTIONS ON IMAGE PROCESSING,
, 2006
"... Probabilistic models of image statistics underlie many approaches in image analysis and processing. An important class of such models have variables whose dependency graph is a tree. If the hidden variables take values on a finite set, most computations with the model can be performed exactly, inclu ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
Probabilistic models of image statistics underlie many approaches in image analysis and processing. An important class of such models have variables whose dependency graph is a tree. If the hidden variables take values on a finite set, most computations with the model can be performed exactly, including the likelihood calculation, training with the EM algorithm, etc. Crouse et al. developed one such model, the hidden Markov tree (HMT). They took particular care to limit the complexity of their model. We argue that it is beneficial to allow more complex treestructured models, describe the use of information theoretic penalties to choose the model complexity, and present experimental results to support these proposals. For these experiments, we use what we call the hierarchical image probability (HIP) model. The differences between the HIP and the HMT models include the use of multivariate Gaussians to model the distributions of local vectors of wavelet coefficients and the use of different numbers of hidden states at each resolution. We demonstrate the broad utility of image distributions by applying the HIP model to classification, synthesis, and compression, across a variety of image types, namely, electrooptical, synthetic aperture radar, and mammograms (digitized Xrays). In all cases, we compare with the HMT.