Jointly Learning the Discriminative Dictionary and Projection for Face Recognition

Algorithm 1

The algorithm of JLDDP.

(1)	Input: the training set , iteration number , parameters , , and .
(2)	Initialize: projection matrix , structured dictionary , .
(3)	Repeat steps 3–6 until convergence or conditions.
(4)	Update with fixed and by equation (7).
(5)	Update with fixed and by equation (8).
(6)	Update with fixed and by equation (12).
(7)	Output: projection matrix P, structured dictionary D, coding coefficient matrix X.