A new normalization technique for cursive handwritten words
This paper presents new techniques for slant and slope removal in cursive handwritten words. Both new methods make no use of heuristics and no manual parameter tuning is needed. This avoids the heavy experimental effort required to find the optimal configuration of a parameter set. A comparison between the new deslanting technique and the method proposed by Bozinovic and Srihari was made by measuring the performance of a word recognition system on different databases. The new technique is shown to improve the recognition rate of the system and to avoid the long exploration of the parameter space needed by the other method.
rr00-32.pdf
openaccess
262.12 KB
Adobe PDF
9416130fb3d384b9b3dcc4cc0a19e3a0