A Novel Compression Method Based on Bandlet and Compressive Sensing for Ultrasound Image

A novel method of compressive sensing based on Bandlet transformation for the compression of ultrasound image is proposed in the paper. Firstly, Bandlet transformation is able to obtain the sparse representation adaptively according to the geometric regularity of the image, it is used to transform the ultrasound image to get the sparse representation in Bandlet domain. Then random measure matrix incoherent with Bandlet basis is used to reduce the data amount of image compression. Finally, matching pursuit algorithm is utilized to reconstruct the ultrasound image from the compressed data. The experiment results show that the proposed method can not only reduce the transmission bandwidth and data amount, also achieve a high quality of ultrasound image reconstruction, which is suitable for ultrasound reconstruction.