Two-channel data gluing technology can extend the dynamic detection range of lidar system effectively and improve the signal to noise ratio of atmospheric echo signal. A lidar data gluing technology based on the non-dominated sorting genetic algorithm II (NSGA-II) and neighborhood rough set (NRS) is proposed to improve the accuracy and stability of data gluing. Three evaluation functions determining goodness of fitting are taken as objective functions and the Pareto optimal solution set is acquired by NSGA-II. The weight which is acquired by NRS is applied to accomplishing the global random searching for the best fitting range. The experimental results show that the proposed algorithm has a favorable gluing result and works steadily in gluing tasks all day.