Improving the Segmentation of Cursive Handwritten Words using Ligature Detection and Neural Validation

@inproceedings{ChengImprovingTS,
  title={Improving the Segmentation of Cursive Handwritten Words using Ligature Detection and Neural Validation},
  author={Chun Ki Cheng and Michael Blumenstein}
}
This paper describes an enhanced neural network-based segmentation technique for improving the segmentation process in cursive handwriting recognition. The technique has two major steps: First, Prospective Segmentation Points (PSPs) are found by an Enhanced Heuristic Segmenter (EHS). EHS locates possible segmentation points based on the ligature and global characteristics of the handwriting. The second part of the technique fuses confidence values obtained from the left and centre character… CONTINUE READING
Highly Cited
This paper has 21 citations. REVIEW CITATIONS