Rapid magnetic resonance imaging (MRI) method based on CS ( compressed sensing ) technique

The invention discloses a rapid magnetic resonance imaging method based on a CS (compressed sensing) technique. The traditional imaging method has relatively low speed and high hardware cost. The method comprises the following steps of: firstly, acquiring variable-density random k spatial data; specifically, determining under sampling rate according to the structural sparsity of an image; carrying out sparse acquisition in a k-space central area and carrying out random sparse acquisition in a k-space peripheral area according to the under sampling rate by combining with the k-space energy distribution rule to generate a variable-density random data acquisition path; acquiring the data according to the determined data acquisition path; then carrying out sparse conversion on an MRI image; and finally, nonlinearly optimizing and reconstructing the image based on Li norm minimum. The method breaks through the limit of the classical Nyguist sampling theorem, accurately reconstructs the signal of the MRI image through randomly acquiring few data points by utilizing a nonlinear optimization algorithm and greatly shortens the data acquisition time.