Most existing regression based classification methods for robust face recognition usually characterize the representation error using L-1-norm or Frobenius-norm for the pixel-level noise or nuclear norm for the image-level noise, and code the coefficients vector by l(1-) norm or l(2)-norm. To our best knowledge, nuclear norm can be used to describe the low rank structural information but may lead to the suboptimal solution, while l(1)-norm or l(2)-norm can promote the sparsity or cooperativity but may neglect the prior information (e.g., the locality and similarity relationship) among data. To solve these drawbacks, we propose two weighted sparse coding regularized nonconvex matrix regression models including weighted parse coding regularized matrix gamma-norm based matrix regression (WS gamma MR) for the structural noise and weighted parse coding regularized matrix gamma-norm plus minimax concave plus (MCP) function based matrix regression (WS gamma(MR)-R-2) for the mixed noise (e.g, structural noise plus sparse noise). The MCP induced nonconvex function can overcome the imbalanced penalization of different singular values and entries of the error image matrix, and the weighted sparse coding can consider the prior information by borrowing a novel distance metric. The variants of inexact augmented Lagrange multiplier (iALM) algorithm including nonconvex iALM (NCiALM) and majorization-minimization iALM (MMiALM) are developed to solve the proposed models, respectively. The matrix gamma-norm based classifier is devised for classification. Finally, experiments on four popular face image databases can validate the superiority of our methods compared with the-state-of-the-art regression methods.