Evolutionary programming(EP) has been applied with success to many numerical and combinatorial optimization problems in recent years. EP has rather slow convergence rates, however, on some function optimization problems. In this paper the Bi-Group evolutionary programming is proposed to overcome the premature convergence. There are two groups in the Bi-Group evolutionary programming. The global group is responsible for searching the whole space. The local group is responsible for searching the local part in detail. The cooperation and specialization between different groups are considered during the algorithm design. The experimental results show the Bi-Group evolutionary programming is efficient in image processing.