Research Article
Dimension Selection and Compression Reconstruction Algorithm of Measurement Matrix Based on Edge Density
@INPROCEEDINGS{10.1007/978-3-030-21730-3_10, author={Jiayin Yu and Erfu Wang}, title={Dimension Selection and Compression Reconstruction Algorithm of Measurement Matrix Based on Edge Density}, proceedings={Green Energy and Networking. 6th EAI International Conference, GreeNets 2019, Dalian, China, May 4, 2019, Proceedings}, proceedings_a={GREENETS}, year={2019}, month={6}, keywords={Compressive sensing Sparse representation Measurement matrix Similarity coefficient}, doi={10.1007/978-3-030-21730-3_10} }- Jiayin Yu
Erfu Wang
Year: 2019
Dimension Selection and Compression Reconstruction Algorithm of Measurement Matrix Based on Edge Density
GREENETS
Springer
DOI: 10.1007/978-3-030-21730-3_10
Abstract
This paper is based on the sparse representation of signals in orthogonal space. Data collection and compressed are combined by compressed sensing theory. The image signal can be reconstructed by fewer observations which we obtained it under the measurement matrix. Compressive sensing theory breaks through the limitation of data sampling. In the theory of compressed sensing, the selection of measurement matrix plays a key role in whether the compressed signal can be reconstructed or not. In this paper, different measurement matrices are selected to achieve the compressive sensing and their similarity coefficient matrices are analyzed to compare the different performance. This paper focus on the coefficient selection of random measurement matrix. To find the relationship between the image structure similarity coefficient and the other characteristic indexes. An algorithm for dimension design of measurement matrix is proposed. A high performance algorithm for image compression perception and image restoration is implemented.