Probabilistic and stochastic approaches can facilitate the search for local and global optima. Evolutionary algorithms, such as genetic population (Jomier et al., 2006; Rivest-Henault et al., 2012; Ruijters et al., 2009), are considered as a strategy that is “less likely to get stuck in a local optimum” (Ruijters et al., 2009). A cost function consisting of the “sum of the Gaussian-blurred intensity values in the [DSA] at the projected model points” (Jomier et al., 2006) is optimized using a genetic algorithm optimizer. Other authors “use the Condensation form of sequential Monte Carlo sampling to estimate a cost function gradient” (Florin et al., 2005) for finding the global minimum. Besides, the Kalman filter is successfully adopted (Curwen et al., 1994; Feldmar et al., 1997; Toledo et al., 1998).
