In this section, WBDL algorithm was evaluated on three classification tasks of simulation experiment, face recognition, and object recognition.
Compared to general dictionary learning algorithm, WBDL model introduces block structure and weight matrix.
WBDL is the algorithm adding weight vector and block structure, simultaneously.
In this section, WBDL algorithm is evaluated through face recognition task on AR face database .
WBDL algorithm is compared with several recently proposed algorithms including SRC , KSVD , D-KSVD , and LC-KSVD .
As shown in Figure 4, after 10 runs, the objective function values decrease very lowly, so WBDL algorithm converges much faster.
In this section, we evaluate WBDL algorithm with existing dictionary learning methods on Extended Yale B face database .
The proposed WBDL algorithm is compared with several recently proposed algorithms including SRC , KSVD , D-KSVD , LC-KSVD , Pl2/1 , and SVGDL .
Finally, we trained class dictionary and learned classifier on the final spatial pyramid features using WBDL algorithm.
As shown in Figure 8, WBDL algorithm maintains the highest classification accuracy in all dictionary size compared with other six methods.
This WBDL dictionary is the product of proto dictionary and corresponding weight vector.
Caption: Figure 2: Performance comparisons of sparse coefficients from BDL model and from WBDL model.