Localization of Shapes Using Statistical Models and Stochastic Optimization

• Published in: IEEE transactions on pattern analysis and machine intelligence 2007-09, Vol.29 (9), p.1603-1615
• Main Authors: Destrempes, F., Mignotte, M., Angers, J.-F.
• Format: Article
• Language: English
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; Computer science; control theory; systems; Computer Simulation; Data Interpretation, Statistical; Deformable models; Exact sciences and technology; Exploration/Selection (E/S) algorithm; Ice; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Image segmentation; Imaging, Three-Dimensional - methods; Iterative algorithms; Localization; Mathematical analysis; Mathematical models; Models, Statistical; Parameter estimation; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Position (location); Principal component analysis; Probabilistic Principal Component Analysis (PPCA); Reproducibility of Results; Segmentation; Sensitivity and Specificity; Shape; Shape localization; Statistical analysis; Statistical distributions; statistical model; stochastic optimization; Stochastic Processes; Stochasticity; Studies; Vectors (mathematics)
• ISSN: 0162-8828; 1939-3539
• DOI: 10.1109/TPAMI.2007.1157

In this paper, we present a new model for deformations of shapes. A pseudolikelihood is based on the statistical distribution of the gradient vector field of the gray level. The prior distribution is based on the probabilistic principal component analysis (PPCA). We also propose a new model based on mixtures of PPCA that is useful in the case of greater variability in the shape. A criterion of global or local object specificity based on a preliminary color segmentation of the image is included into the model. The localization of a shape in an image is then viewed as minimizing the corresponding Gibbs field. We use the exploration/selection (E/S) stochastic algorithm in order to find the optimal deformation. This yields a new unsupervised statistical method for localization of shapes. In order to estimate the statistical parameters for the gradient vector field of the gray level, we use an iterative conditional estimation (ICE) procedure. The color segmentation of the image can be computed with an exploration/selection/estimation (ESE) procedure.