Description

• Synthetic and real MOD can only be implemented for the effectiveness of the proposed scheme, also in comparison with other sampling techniques. • Global voting algorithm is performed, based on local density and trajectory similarity information. • Propose a method for trajectory segmentation and sampling based on the representativeness of the (sub)trajectories in the MOD. • Depends on the partitioned data, sampling can be proceed for accuracy result. • For each segment of the trajectory, forming a local trajectory descriptor that represents line segment representativeness. • Sequence of this descriptor over a trajectory gives the voting signal of the trajectory, where high values correspond to the most representative parts.