Text fragment recovery method based on discrete artificial bee colony algorithm

This paper proposes a new method based on discrete artificial bee colony algorithm to solve the problem of text fragment recovery. First, text fragments are clustered by row according to the distribution of the characters in the fragments. Subsequently, the matching degree of the edge information of each row of scrap paper is used as an index to measure whether the debris is adjacent to the left and right. An optimal restoration model with an overall matching degree is established. Finally, this paper uses discrete ABC algorithm to solve the problem. The experimental results show that the algorithm has good stability, efficiency and robustness in the restoration of text fragments.