文献类型：期刊文献

英文题名：Quaternion discrete LOTHAD transforms

作者：Zhu, Hongqing[1]; Zhu, Xiaoli[1]; Li, Qian[1]

机构：[1] School of Information Science and Engineering, East China University of Science and Technology, Shanghai , China

年份：2014

卷号：11

期号：15

起止页码：5527

外文期刊名：Journal of Information and Computational Science

收录：EI(收录号：20144400136080)

语种：英文

外文关键词：Hadamard matrices - Hadamard transforms

摘要：This paper first introduces a new orthogonal transform known as LOTHAD. It results from the Kronecker product of the lapped orthogonal transform and the discrete Hadamard transform. Using the algebra of quaternions, the study presents the quaternion COSHAD and LOTHAD transforms to address colour images in a holistic manner. Then, the paper provides the theoretical framework to construct the blockbased Compressed Sensing (CS) colour image description methods and to compare the performance of the proposed approaches with recent state-of-the-art CS methods. The results show that the presented methods achieve significant improvement than other methods in terms of the degree of sparsity and the reconstruction quality. Copyright ? 2014 Binary Information Press.