Precise Recognition of Blurred Chinese Characters by Considering Change in Distribution
Shinichiro Omachi, Fang Hong Sun · 2000
In this paper, a newalgorithm for Chinese character recognition is presented. This misv d has an especial e#ect on characterimact those are not clear.Comr.v with a clear characterimact a blurred character imra always has a big change in shape thatma es the recognitionmec di#cult. In order to recognize blurred characters precisely, the concept of degree of blur isem9 yed. We also in estigate how the distribution of feature ectors changes if a part of characterimac is crashed. Since the degree of blur is calculatedfrom each sub-area of characterimact the condition of theimv can be described clearly. The distribution can be reconstructed to respond to the change expressed by the degree of blur. Themev d proposed in this paper can recognize blurred characters precisely by m difying the Mahalanobis distance function according to change in distribution. The e#ecti eness of the newmwv d is shown by experim ts with blurred characterimacte 1