A survey on handwriting recognition based on the trajectory recovery technique

The success of online recognition systems leads to improving the performance of offline recognition systems using the recovery of dynamic information. Therefore, it appears the techniques that extract dynamic information from static text. Several techniques have been proposed in the field of trajectory recovery. The principle of these techniques is to find an oriented path similar to that used by the writer in the writing operation. Many works have proved that the results of handwriting recognition systems can be improved if the temporal order of original trajectory will be restored correctly. The aim of this study is as follow. Firstly, we introduce the principle of the trajectory recovery approach and the basic concepts of its processes. Secondly, we give a bibliography of trajectory recovery-based handwriting recognition system. We consider our work as the first survey that focuses on trajectory recovery-based handwriting recognition system.