To solve the problem of low population diversity and weak exploitation of sparrow search algorithm, an improved sparrow search algorithm based on orthogonal-opposition-based learning (OOLSSA) is proposed in this paper. First, a normal mutation operator is used to enrich the diversity of algorithm population. Second, the opposition-based learning is used to enhance the ability of the algorithm to jump out of local optimum. Then, orthogonal-opposition-based learning is introduced after the update of the scrounger position to accelerate the convergence of the algorithm. Finally, performance test based on fifteen benchmark test functions, non-parametric Friedman test and balance analysis of algorithms shows that compared with six traditional optimization algorithms and two improved algorithms, OOLSSA has better searching performance on exploration and exploitation ability and convergence speed.