Abstract
Abstract. In this paper we propose a new method for the automatic extraction of tidal channels in digital terrain models (DTM) using a sampling approach based on marked point processes. In our model, the tidal channel system is represented by an undirected, acyclic graph. The graph is iteratively generated and fitted to the data using stochastic optimization based on a Reversible Jump Markov Chain Monte Carlo (RJMCMC) sampler and simulated annealing. The nodes of the graph represent junction points of the channel system and the edges straight line segments with a certain width in between. In each sampling step, the current configuration of nodes and edges is modified. The changes are accepted or rejected depending on the probability density function for the configuration which evaluates the conformity of the current status with a pre-defined model for tidal channels. In this model we favour high DTM gradient magnitudes at the edge borders and penalize a graph configuration consisting of non-connected components, overlapping segments and edges with atypical intersection angles. We present the method of our graph based model and show results for lidar data, which serve of a proof of concept of our approach.
Highlights
Strategies for the automatic detection of objects from images can be divided into top-down and bottom-up approaches
In the parts marked by circles (Figure 4, bottom) the graph is not connected. Another good result is that segments only connected to one node are completely eliminated during the sampling process
If we reduce c from 200 to 100, some of the smaller tidal channels are detected in the upper part of the scene (Fig 5)
Summary
Strategies for the automatic detection of objects from images can be divided into top-down and bottom-up approaches. Top-down approaches integrate knowledge about the objects and search for matches with the input data Stochastic methods such as marked point processes (Daley and Vere-Jones, 2003), which achieve good results in various disciplines, belong to the top-down approaches. Model knowledge can be integrated in different ways To this end Lafarge et al (2010) developed a flexible approach for the detection of different kinds of objects in images. The objects are modelled by rectangles whose conformity with the input data is evaluated based on the gray values of the pixels inside the rectangles and in their local neighbourhood. Road networks are extracted in the approach of Lacoste et al (2005) This is done by taking into account different types of relations between segments and evaluating their connectivity and orientation. Extreme closeness of objects is penalized by a constant parameter
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
