Minimum-Cost Path Algorithm for Separating Touching Characters

D.-R. Lee and I.-S. Oh (Korea)


Optical character recognition (OCR), touching characters, document images, minimum-cost path, greedy algorithm


The paper proposes an algorithm which finds a nonlinear cut path for a greyscale touching character image. It proposes a new penalizing rule which is probable to guide correctly the cut path for touching situations difficult to separate. Additionally the algorithm adopts a strategy of producing both the downward and upward paths and selecting better one. The experimental results on actual touching character images showed that the proposed algorithm was superior to conventional algorithms.

Important Links:

Go Back