Infrared Small Dim Target detection Based on Weighted Nuclear Norm Minimization

摘要

Small dim target detection is a crucial technique of Infrared Search and Track Systems. Image patch model (IPI) based small dim target detection method, which factorizes a data matrix into a low rank background matrix and a sparse target matrix, achieves a good performance in complex scenery. Yet, equally weighted singular value in this model cannot flexibly describe more complex background. To cope with the problem, this paper proposes a weight nuclear norm minimization based IPI model, which considers each singular value differently. We give a optimization method to solve the model. Furthermore, we present a target extraction technique from sparse target image robustly. Experimental results on a comprehensive infrared target image dataset show our method performs better than IPI model based method as well as other conventional approaches.