An Efficient Segmentation Based Recognition of Unconstrained Handwritten Touching Digits


The KIPS Transactions:PartB , Vol. 8, No. 3, pp. 223-230, Jun. 2001
10.3745/KIPSTB.2001.8.3.223,   PDF Download:

Abstract

This paper proposes an efficient segmentation-based recognition algorithm for unconstrained handwritten touching digits. We classify touching into six types based on an analysis of the ligature component and the characteristic of candidate segmentation points. Four kinds of candidate segmentation points are obtained from contour profiles and the final segmentation point of touching digits is determined by verifying the candidate segment combinations. The main advantages of this method are that reliable segment combinations are used in the multiple hypothesis recognition, and segmentation errors of traditional segmentation-based approach is reduced by introducing prioritized segmentation points. To evaluate the proposed method, we have experimented with 3,500 touching digits of the NIST database. An encouraging recognition rate of 92.5% has been obtained.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
K. K. Kim, J. H. Kim, H. J. Park, K. D. Bu, "An Efficient Segmentation Based Recognition of Unconstrained Handwritten Touching Digits," The KIPS Transactions:PartB , vol. 8, no. 3, pp. 223-230, 2001. DOI: 10.3745/KIPSTB.2001.8.3.223.

[ACM Style]
Kye Kyung Kim, Jin Ho Kim, Hee Joo Park, and Ki Dong Bu. 2001. An Efficient Segmentation Based Recognition of Unconstrained Handwritten Touching Digits. The KIPS Transactions:PartB , 8, 3, (2001), 223-230. DOI: 10.3745/KIPSTB.2001.8.3.223.