Techniques for static handwriting trajectory recovery: a survey
DAS '10 Proceedings of the 9th IAPR International Workshop on Document Analysis Systems
| Hi-index | 0.00 |
Summary form only given. In this paper, we present a system of restoration of temporal order in the offline Arabic handwritten tracing. The word image, captured in level of gray from a scanner with a resolution of 300 dpi, passes by four stages of preprocessing: binarization, filtering, smoothing and elimination of the diacritical signs. A first algorithm calculates for every point of the contour a curvature function. The local maximas of this function correspond to the dominant points of the contour. The dominant points correspond to all angular variations on the contour (end of the stroke, cross point, branch point, pocket, buckle...). A sweep of a window of a dimension equal to the triple of the thickness, of the stroke permits to detect cross and branch points of features and a suitable algorithm permits to solve the problem of the temporal order in these ambiguous zones. Finally, another algorithm permits to detect the starting point and to follow the tracing in order to reconstitute the order in which it has been written. The system is improved by an algorithm that treats the particular cases concerning the starting point, the ambiguous zone analysis and the progression of the system at the time of the follow-up of the contour. Preliminary results obtained from our basis (50 words written by 2 people), are: detection of the point of departure: 96%; correct interpretation of ambiguous zones: 95%; and movement of pen (temporal order) recovered: 92%.