Hierarchy in Picture Segmentation: A Stepwise Optimization Approach
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Bayesian Approach to Reconstruction from Incomplete Projections of a Multiple Object 3D Domain
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Electric current computed tomography eigenvalues
SIAM Journal on Applied Mathematics
SIAM Journal on Applied Mathematics
Wavelet-like bases for the fast solutions of second-kind integral equations
SIAM Journal on Scientific Computing
SIAM Journal on Applied Mathematics
Estimating target location from scattered field data
Signal Processing
Picture Segmentation by a Tree Traversal Algorithm
Journal of the ACM (JACM)
Detection, Estimation, and Modulation Theory: Radar-Sonar Signal Processing and Gaussian Signals in Noise
ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 1)-Volume 1 - Volume 1
Correlation structure of the discrete wavelet coefficients of fractional Brownian motion
IEEE Transactions on Information Theory - Part 2
Wavelet analysis and synthesis of fractional Brownian motion
IEEE Transactions on Information Theory - Part 2
Hi-index | 0.01 |
Inthis paper we explore the utility of multiscale and statisticaltechniques for detecting and characterizing the structure oflocalized anomalies in a medium based upon observations of scatteredenergy obtained at the boundaries of the region of interest.Wavelet transform techniques are used to provide an efficientand physically meaningful method for modeling the non-anomalousstructure of the medium under investigation. We employ decision-theoreticmethods both to analyze a variety of difficulties associatedwith the anomaly detection problem and as the basis for an algorithmto perform anomaly detection and estimation. These methods allowfor a quantitative evaluation of the manner in which the performanceof the algorithms is impacted by the amplitudes, spatial sizes,and positions of anomalous areas in the overall region of interest.Given the insight provided by this work, we formulate and analyzean algorithm for determining the number, location, and magnitudesassociated with a set of anomaly structures. This approach isbased upon the use of a Generalized, M-ary Likelihood Ratio Testto successively subdivide the region as a means of localizinganomalous areas in both space and scale. Examples of our multiscaleinversion algorithm are presented using the Born approximationof an electrical conductivity problem formulated so as to illustratemany of the features associated with similar detection problemsarising in fields such as geophysical prospecting, ultrasonicimaging, and medical imaging.