Table 1.

The NSCR-based classifier.

The NSCR based classifier
1	Input: training sample matrix X = [X1, ⋯, Xk] and query sample y
2	Normalize each column of matrix X and query sample y to the unit L_2 norm
3	The encoding vector of y on X is solved by the NSCR model
4	Calculate the coefficient matrix: $\widehat{c}=\arg {\min }_c‖y^{\prime }=Xc‖+\alpha {‖c‖}_2^2+\beta cs.t.c\ge 0$
5	Calculate residual similarity: $r_k={‖y-X_k{\widehat{c}}_k‖}_2$
6	Output label category: label(y) = arg min{rk}