The statistics of natural images play an important role in many image processing tasks. In particular, statistical assumptions about differences between neighboring pixel values are used extensively in the form of pri...
详细信息
ISBN:
(纸本)0819461059
The statistics of natural images play an important role in many image processing tasks. In particular, statistical assumptions about differences between neighboring pixel values are used extensively in the form of prior information for many diverse applications. The most common assumption is that these pixel difference values can be described be either a Laplace or Generalized Gaussian distribution. The statistical validity of these two assumptions is investigated formally in this paper by means of Chi-squared goodness of fit tests. The Laplace and Generalized Gaussian distributions are seen to deviate from real images, with the main source of error being the large number of zero and close to zero nearby pixel difference values. These values correspond to the relatively uniform areas of the image. A mixture distribution is proposed to retain the edge modeling ability of the Laplace or Generalized Gaussian distribution, and to improve the modeling of the effects introduced by smooth image regions. The Chi-squared tests of fit indicate that the mixture distribution offers a significant improvement in fit.
An information-theoretic method is described for automatically determining the best number of clusters. it is motivated by Rissanen's minimum description length principle that states the best representation is the...
详细信息
ISBN:
(纸本)0819462918
An information-theoretic method is described for automatically determining the best number of clusters. it is motivated by Rissanen's minimum description length principle that states the best representation is the one with the fewest bits. The method is evaluated using two different clustering algorithms: a mode finder based on scale-space algorithm, and a vector quantizer (VQ). Synthetic, single- and multi-band image clustering examples are presented. Clusterings produced by the mode finder are shown to better correspond to distinguishable surface categories in the scene than those produced by the VQ algorithm. VQ clusterings are evaluated within an anomaly detector, which detects manmade object/changes as spectral outliers within a set of background clusters. It is shown that the optimal VQ clustering (the one with the fewest bits) produces the best detection performance.
Breast cancer continues to be the most common malignancy of women in the United States. Nuclear imaging techniques such as positron emission tomography (PET) have been widely used for the staging of cancer. The primar...
详细信息
ISBN:
(纸本)0819461059
Breast cancer continues to be the most common malignancy of women in the United States. Nuclear imaging techniques such as positron emission tomography (PET) have been widely used for the staging of cancer. The primary limitations of PET for breast cancer diagnosis are the lack of a highly specific radiotracer and the limited
resolution of imaging systems. The sensitivity for detecting small lesions is very low. Many groups are developing positron emission mammography (PEM) systems dedicated for breast imaging using high resolution detectors. Although image resolution is significantly improved compared to whole-body PET systems, the clinical value of
a PEM system is yet to be proven,3.4 Most PET systems have limitations in imaging tissues near the chest walls and lymph nodes. The proposed system addresses the sampling requirements specific to breast imaging and achieves high resolution in PET images of breast and thorax.
The 2D segmentation method CSC (Color Structure Code) for color images has recently been generalized to 3D color or grey valued images. To apply this technique for an automated analysis of 3D MR brain images a few pre...
详细信息
ISBN:
(纸本)0819464236
The 2D segmentation method CSC (Color Structure Code) for color images has recently been generalized to 3D color or grey valued images. To apply this technique for an automated analysis of 3D MR brain images a few preprocessing and postprocessing steps have been added. We present this new brain analysis technique and compare it with SPM.
The hidden Markov tree (HMT) model is a novel statistical model for image processing on wavelet domain. The HMT model captures the persistence property of wavelet coefficients, but lacks the clustering property of wav...
详细信息
The hidden Markov tree (HMT) model is a novel statistical model for image processing on wavelet domain. The HMT model captures the persistence property of wavelet coefficients, but lacks the clustering property of wavelet coefficients within a scale. We propose the contextual hidden Markov tree (CHMT) model to enhance the clustering property of the HMT model by adding extended coefficients associated with wavelet coefficients. The extended coefficients do not change the wavelet tree structure but enhance the intrascale dependencies of the HMT model. Hence, the training scheme of the HMT model can be modified to estimate the parameters of the CHMT model. In experiments, the proposed CHMT model produces almost better results than the HMT model produces for image denoising. Furthermore, the CHMT model requires fewer iterations of training than the HMT model to achieve the same denoised results. (c) 2005 SPIE and IS&T.
When recording three-dimensional (3D) images by the method of optical sectioning microscopy, each optical section contains the in-focus information plus out-of-focus contributions that obscure the in-focus detail and ...
详细信息
ISBN:
(纸本)0819456756
When recording three-dimensional (3D) images by the method of optical sectioning microscopy, each optical section contains the in-focus information plus out-of-focus contributions that obscure the in-focus detail and reduce contrast. There are several methods to remove or prevent the out-of-focus contributions from the stack of optical sections. One such method is image estimation -the use of a computer program based on a mathematical description of the microscope to remove the out-of-focus contributions. Another method is the use of structured illumination and a simple arithmetic operation to obtain a image that in which the out-of-focus contributions are greatly reduced. We derived a method for image estimation that uses the images collected from the structured-illumination microscope. The method improves the resolution of small detail over that possible with the structured illumination using the simple arithmetic formula.
作者：
Avesta, NAboulnasr, TTurku Univ
Dept Informat Technol Lab Electron & Informat Technol FIN-20014 Turku Finland Univ Ottawa
CASP Lab Sch Informat Technol Engn Ottawa ON K1N GN5 Canada
Adaptive processes often attempt to minimize the mean square error (MSE) to filter a partially observed digital signal. While mathematically tractable, the MSE criterion often causes over-smoothing of the filtered sig...
详细信息
Adaptive processes often attempt to minimize the mean square error (MSE) to filter a partially observed digital signal. While mathematically tractable, the MSE criterion often causes over-smoothing of the filtered signal. In this paper, we propose using maximum entropy (ME) as the optimization criterion to avoid the oversmoothing of signals. This criterion is motivated by the fact that ME methods make no assumptions regarding the unobserved data, aside from explicitly stated ones. The maximum entropy Kalman filter presented in this paper employs ME as its optimization criterion to explicitly identify the appropriate parameters of the standard Kalman filter, for the purpose of image compression and reconstruction. (C) 2004 SPIE and IST.
Lesion detection is an important task in emission tomography. Localization ROC (LROC) studies are often used to analyze the lesion detection and localization performance. Most researchers rely on Monte Carlo reconstru...
详细信息
ISBN:
(纸本)0819452858
Lesion detection is an important task in emission tomography. Localization ROC (LROC) studies are often used to analyze the lesion detection and localization performance. Most researchers rely on Monte Carlo reconstruction samples to obtain LROC curves, which can be very time-consuming for iterative algorithms. In this paper we develop a fast approach to obtain LROC curves that does not require Monte Carlo reconstructions. We use a channelized Hotelling observer model to search for lesions, and the results can be easily extended to other numerical observers. We theoretically analyzed the mean and covariance of the observer output. Assuming the observer outputs are multivariate Gaussian random variables, an LROC curve can be directly generated by integrating the conditional probability density functions. The high-dimensional integrals are calculated using a Monte Carlo method. The proposed approach is very fast because no iterative reconstruction is involved. Computer simulations show that the results of the proposed method match well with those obtained using the tradition LROC analysis.
We have developed a method for clustering features into objects by taking those features which include intensity, orientations and colors from the most salient points in an image as determined by our biologically moti...
详细信息
ISBN:
(纸本)0819451959
We have developed a method for clustering features into objects by taking those features which include intensity, orientations and colors from the most salient points in an image as determined by our biologically motivated saliency program. We can train a program to cluster these features by only supplying as training input the number of objects that should appear in an image. We do this by clustering from a technique that involves linking nodes in a minimum spanning tree by not only distance, but by a density metric as well. We can then form classes over objects or object segmentation in a novel validation set by training over a set of seven soft and hard parameters. We discus as well the uses of such a flexible method in landmark based navigation since a robot using such a method may have a better ability to generalize over the features and objects.
Images recorded in ground areas potentially containing surface laid land mines are considered. The first hypothesis is that the image is of clutter (grass) only, while the alternative is that the image contains a part...
详细信息
ISBN:
(纸本)0819453382
Images recorded in ground areas potentially containing surface laid land mines are considered. The first hypothesis is that the image is of clutter (grass) only, while the alternative is that the image contains a partially occluded (covered) land mine in addition to the clutter. In such a scenario, the occlusion pattern is unknown and has to be treated as a nuisance parameter. In a previous paper it was shown that deterministic treatment of the unknown occlusion pattern, in companion with the applied model, renders a substantial increase in detector performance as compared to employment of the traditional additive model. However, a deterministic assumption ignores possible correlation and additional gains could be possible by taking the spatial properties into account. In order to incorporate knowledge regarding the occlusion, the spatial distribution is characterized in terms of an underlying Markov Random Field (MRF) model. A major concern with MRF models is their complexity. Therefore, in addition to this, a less computationally demanding technique to accommodate the occlusion behavior is also proposed. The main purpose of this paper is to investigate if significant gains are possible by acknowledging the spatial dependence. Evaluation on data using real occluded targets however indicates that the gain seem to be marginal.
暂无评论