Active shape models—their training and application
Computer Vision and Image Understanding
Object Matching Using Deformable Templates
IEEE Transactions on Pattern Analysis and Machine Intelligence
Using Generative Models for Handwritten Digit Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
Shape Matching Using LAT and its Application to Handwritten Numeral Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
Deformation Transformation for Handwritten Chinese Character Shape Correction
ICMI '00 Proceedings of the Third International Conference on Advances in Multimodal Interfaces
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
Deformable models (DMs) generally possess shape-varying capability, making them particularly suitable for extracting and recognizing non-rigid objects. However, most of the existing DMs are limited to model a close or open contour and hence they are not applicable to complex handwriting patterns like Chinese characters and signatures of which structure plays an important role. In this paper, a new type of DMs called structural deformable model (SDM) is proposed and preliminary results are reported. The new model takes structural information into accounts by representing handwriting patterns as a set of active contours that are structurally connected with each other and contain information about the orientation of stroke segments. Appropriate internal and external energy functions are formulated to preserve the model structure and satisfy the data match criterion. By applying the steepest descent method and proposing an effective initialization scheme, the deformation process is derived. The performance of the new model is demonstrated through a small scale Chinese character recognition experiment.