Small target detection in infrared video sequence using robust dictionary learning

Small target detection in infrared video sequence is a challenging problem. In this paper, a collaborative structured sparse coding (SSC) model which incorporates the L1,2 and L2,1 regularization terms is proposed. The Alternating Direction Method of Multiplier (ADMM) is developed to solve this model. Further, online dictionary learning is embedded into the model and temporal information is incorporated to eliminate the clutters and noises. Extensive synthetic and real data experiments show that our method obtains better detection performance than baseline methods and state-of-art infrared-patch-image (IPI) model.