DescriptionIn this thesis, we study an open set recognition algorithm that is based on the Sparse Representation-based Classification (SRC) method. By modeling the tail distributions of the matched and non-matched reconstruction errors using the statistical Extreme Value Theory (EVT), we simplify the open set recognition problem into a set of hy- pothesis testing problems. The confidence scores corresponding to the tail distributions of a novel test sample are then fused to determine its identity. The effectiveness of the proposed method is demonstrated using three publicly available image and object classification datasets and it is shown that this method can perform significantly better than many competitive open set recognition algorithms.