In recent years,swarm intelligence algorithms have made important progress and remarkable achievements in spectral unmixing of hyperspectral image by solving combinatorial optimization or continuous optimization problems.In this paper,the background of the research of spectral unmixing in hyperspectral image and the characteristics of swarm intelligence algorithm were reviewed firstly,and then the optimization model and the spectral mixture model were teased out correspondingly.Then the endmember extraction and abundance inversion method based on swarm intelligent algorithms were introduced.Finally the accuracy of spectral unmixing achieved by swarm intelligence algorithms and other traditional algorithms was evaluated through two experiments.In addition,the advantages and problems of swarm intelligence algorithm in hyperspectral image information extraction were also summarized in this paper.