Sparse Representation over Learned Dictionary for Symbol Recognition

TitleSparse Representation over Learned Dictionary for Symbol Recognition
Publication TypeJournal Article
Year of Publication2016
AuthorsDo, T-H, Tabbone, S, Ramos-Terrades, O
JournalSignal Processing
Volume125
Pagination36-47
Date Published1/2016
Abstract

In this paper we propose an original sparse vector model for symbol retrieval task. More specifically, we apply the K-SVD algorithm for learning a visual dictionary based on symbol descriptors locally computed around interest points. Results on benchmark datasets show that the obtained sparse representation is competitive related to state-of-the-art methods. Moreover, our sparse representation is invariant to rotation and scale transforms and also robust to degraded images and distorted symbols. Thereby, the learned visual dictionary is able to represent instances of unseen classes of symbols.

DOI10.1016/j.sigpro.2015.12.020