Component optimization for image understanding: a Bayesian approach

In this paper, the optimizations of three fundamental components of image understanding: segmentation/annotation, 3D sensing (stereo) and 3D fitting, are posed and integrated within a Bayesian framework. This approach benefits from recent advances in statistical learning which have resulted in greatly improved flexibility and robustness. The first two components produce annotation (region labeling) and depth maps for the input images, while the third module integrates and resolves the inconsistencies between region labels and depth maps to fit most likely 3D models. To illustrate the application of these ideas, we have focused on the difficult problem of fitting individual tree models to tree stands which is a major challenge for vision-based forestry inventory systems.

[1]  D. Scharstein,et al.  A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms , 2001, Proceedings IEEE Workshop on Stereo and Multi-Baseline Vision (SMBV 2001).

[2]  A. Gray,et al.  Modern Differential Geometry of Curves and Surfaces with Mathematica, Third Edition (Studies in Advanced Mathematics) , 2006 .

[3]  F. Gougeon.  A Crown-Following Approach to the Automatic Delineation of Individual Tree Crowns in High Spatial Resolution Aerial Images , 1995 .

[4]  Robert J. Woodham,et al.  The automatic recognition of individual trees in aerial images of forests based on a synthetic tree crown image model , 1996 .

[5]  S. Nielsen.  The stochastic EM algorithm: estimation and asymptotic results , 2000 .

[6]  Gilles Celeux,et al.  On Stochastic Versions of the EM Algorithm , 1995 .

[7]  Axel Pinz,et al.  A computer vision system for the recognition of trees in aerial photographs , 1991 .

[8]  R. Stephenson.  A and V , 1962, The British journal of ophthalmology.

[9]  G. Celeux,et al.  Stochastic versions of the em algorithm: an experimental study in the mixture case , 1996 .

[10]  Nanning Zheng,et al.  Stereo Matching Using Belief Propagation , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Morten Larsen,et al.  Optimizing templates for finding trees in aerial photographs , 1998, Pattern Recognit. Lett..

[12]  Terry Caelli,et al.  Bayesian Stereo Matching , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[13]  Steven W. Zucker,et al.  On the Foundations of Relaxation Labeling Processes , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Hui Cheng,et al.  Multiscale Bayesian segmentation using a trainable context model , 2001, IEEE Trans. Image Process..

[15]  A. Gray.  Modern Differential Geometry of Curves and Surfaces , 1993 .

[16]  D. Leckie,et al.  Forest inventory update in Canada , 1996 .

[17]  Peter Green,et al.  Markov chain Monte Carlo in Practice , 1996 .

[18]  J. Tilton.  Multisource Data Integration in Remote Sensing , 1991 .

[19]  J. Besag.  On the Statistical Analysis of Dirty Pictures , 1986 .

[20]  Christian P. Robert,et al.  The Bayesian choice : from decision-theoretic foundations to computational implementation , 2007 .

[21]  Terry Caelli,et al.  A trainable hierarchical hidden Markov tree model for color image annotation , 2002, Object recognition supported by user interaction for service robots.