Model Based Methods

In paper [44] the classifiaction into car types is solved. Car types are : sedan, pick-up, hatchback, wagon and van. The database had 393 vehicles and only 32 vehicles were misclassified which yields the classification rate of 91%. The paper proposes a segmentation algorithm using deformable template models to segment a vehicle of interest both from the stationary background and other moving vehicles in an image sequence. A polygonal template is defined to characterize a general model of a vehicle and a priory probability density function is derived to constrain the template to be deformed within a set of allowed shapes. The segmentation problem is reduced to a minimization problem and solved by a Metropolis algorithm.