In visual simultaneous localization and mapping (SLAM), the loop closure detection module is used to eliminate pose drift and obtain globally consistent trajectory and map. Aiming to solve the problems of low efficiency and accuracy in traditional SLAM loop closure detection methods, an SLAM loop closure detection method based on the Harris hawks optimization (HHO) algorithm is proposed in this article. First, the features from accelerated segment test (FAST) algorithm is modified to extract image features and generate robust descriptors of image features. Then, the problem of solving the maximum similarity between the current and historical frame images in loop closure detection is transformed into a maximum optimization problem. Finally, the HHO algorithm for solving the loop closure detection problem is obtained by designing individual coding mode and fitness function. The experimental results show that the proposed SLAM loop closure detection method based on HHO algorithm has higher efficiency and accuracy compared with the loop closure detection methods based on bag of word and particle swarm optimization (PSO) algorithm.