In this paper, we propose a joint sparse and collaborative representation-based algorithm for target detection in hyperspectral imagery. The proposed target detection is achieved by the representation of the test samples using a target library and a background library. The sparse representation of given target samples is solved by an ℓ1-norm minimization of the representation weight vector, and the collaborative representation of background samples is estimated by an ℓ2-norm minimization. The detection output of the test sample is determined by the difference between sparse reconstruction and collaborative reconstruction. Experimental results show that this algorithm outperforms the existing hyperspectral target detection algorithms, such as adaptive coherence estimator and pure sparse representation-based detector.