A Novel Camouflage Face Recognition Method Based on Joint Sparsity Model

摘要

Many classic and contemporary face recognition algorithms work well under normal conditions, but degrade sharply when they are used at identifying an individual with an altered appearance due to makeup, or more severely, camouflage. We propose a face recognition algorithm based on Joint Sparsity Model (JSM) with Principal Component Analysis (PCA) that achieves a high degree of robustness and stability to camouflage, which extracts a common component, an innovation component from an ensemble of face images. These components have their respective physical significance in terms of representing different types of information in the original ensemble, hence facilitating analysis task such as face recognition. Experimental results on AR Face Database show that the JSM-PCA can improve recognition rate and speed compared with Sparse Representation Classification (SRC) algorithm. The result of JSM-PCA can generalize well to the face recognition, even with a few training image per class.