Inspired by the classical segmentation algorithm, this paper proposes a grayscale image segmentation algorithm based on the image complexity. Firstly, the grayscale image is divided into background and target categories by the threshold, the corresponding histogram and total number of pixels are calculated, as well as the complexity of objects. Secondly, according to the image complexity segmentation criterion, the image complexity of each gray level is calculated. Finally, the optimal segmentation threshold is obtained by the minimum value of the object complexity. Compared with the other three classical algorithms, the experimental results show that the proposed image segmentation algorithm is fast, stable and efficient.