Title: MULTI-STAGE OMP sparse coding using local matching pursuit atoms selection
Abstract:A new multi-stage approach based on component extraction is proposed to more efficiently address the sparse representation problem. In each stage a pre-set number of coefficients are chosen for recons...A new multi-stage approach based on component extraction is proposed to more efficiently address the sparse representation problem. In each stage a pre-set number of coefficients are chosen for reconstructing each signal component. A global search is performed to extract a lower dimensional sub-dictionary consisting of a sorted set of candidate atoms to represent the signal component, corresponding to the stage. The best representing atoms are then selected from the sub dictionary using the Matching Pursuit (MP) method. Afterwards, the sparse coefficients are updated in the same manner in which the Orthogonal Matching Pursuit (OMP) operates. The proposed method is more efficient that the conventional OMP methods. To evaluate the performance of the proposed method, it is compared to OMP and Stagewise OMP (StOMP), which are conceptually the most similar to the proposed approach. The results illustrate the proposed method is more time efficient than OMP and more robust and sparser than the StOMP.Read More
Publication Year: 2013
Publication Date: 2013-05-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
Cited By Count: 9
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot