next up previous
Next: Results for thin network Up: Roads Extraction using a Previous: Data model for extracting

Construction of a RJMCMC dynamics

 
The different movements defining the proposal



\includegraphics[width=10cm]{/u/mahler/0/ariana/xdescomb/PAPER/icip99/tree.ps}


 
Simulation of Candy model



RJMCMC dynamics : adding, deleting, moving segments in the image. The movements (the transition kernels) need to guarantee :

Some movements :

$\cdot$ birth and death of a free segment
$\cdot$ birth and death of a simple connected segment
$\cdot$ birth and death of a double connected segment
$\cdot$ modifying the orientation of a single or simple connected segment
$\cdot$ modifying the length of a single or simple connected segment
$\cdot$ modifying the position of a single connected segment
$\cdot$ modifying the position of a simple connected segment


 
Example of the acceptance ratio



Birth and death of a free segment : $S^{\prime}=S \cup \zeta$.

Let be Pn the probability to choose the birth of a free segment and Pm the probability to choose to kill one.

The acceptance ratio is :

\begin{displaymath}R=\frac {f(S^{\prime})Q(S^{\prime} \rightarrow S)}{f(S)Q(S \rightarrow S^{\prime})}
\end{displaymath} (22)

We have :

\begin{displaymath}Q(S^{\prime} \rightarrow S)=P_{m} \times \frac{1}{n_{d}+1}
\end{displaymath} (23)

nd is the number of free segments in the configuration S.

And :

\begin{displaymath}Q(S \rightarrow S^{\prime})=P_{n} \times \frac{1}{\nu(T)} \times
\frac{1}{2\pi(l_{max}-l_{min})(w_{max}-w_{min})}
\end{displaymath} (24)

We obtain :

\begin{displaymath}R=\frac{P_{m}}{P_{n}} \times \frac{2\pi\nu(T)(l_{max}-l_{min})(w_{max}-w_{min})}{n_{d}+1} \times \frac{f(S \cup \zeta)}{f(S)}
\end{displaymath} (25)


 
Realizations of the Candy model




 
Figure: Realizations of the prior model with different densities : a) $\beta = 0.25$ b) $\beta =0.5$
a) \includegraphics[width=8cm]{/u/biotite/0/ariana/rstoica/DOCS99/FIGURES/PRIOR/prior_25.eps}
b) \includegraphics[width=8cm]{/u/biotite/0/ariana/rstoica/DOCS99/FIGURES/PRIOR/prior_50.eps}

Statistics of the Candy model




 
Figure 10: Statistics of the segments
a) \includegraphics[width=7cm]{/u/biotite/0/ariana/rstoica/DOCS99/FIGURES/PRIOR/s_t_25.eps} b) \includegraphics[width=7cm]{/u/biotite/0/ariana/rstoica/DOCS99/FIGURES/PRIOR/s_l_25.eps}
c) \includegraphics[width=7cm]{/u/biotite/0/ariana/rstoica/DOCS99/FIGURES/PRIOR/s_1_25.eps} d) \includegraphics[width=7cm]{/u/biotite/0/ariana/rstoica/DOCS99/FIGURES/PRIOR/s_2_25.eps}


next up previous
Next: Results for thin network Up: Roads Extraction using a Previous: Data model for extracting
Radu Stoica
2000-04-17