Magnetic resonance image denoising method based on sparse dictionary learning

• Main Authors: YAO ZHONGMING, LUO YIQI, FAN ZIJIA, GUO SHANGHUI, XIN JUNCHANG, WU JIAN, WANG ZHONGYANG, WANG ZHIQIONG
• Format: Patent
• Language: Chinese; English
• Published: 26.11.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention provides a magnetic resonance image denoising method based on sparse dictionary learning, and relates to the technical field of computer aided diagnosis. The method comprises the following steps: firstly, constructing a total variation dictionary denoising model, reconstructing a magnetic resonance image and correcting the magnetic resonance image by applying a TV norm; then, carrying out sparse coding and sparse dictionary updating, using a gradient descent straight line search method for carrying out optimal solution on the target functional of the sparse dictionary D, and achieving updating of the sparse dictionary; and constructing an adaptive atom dictionary learning model, performing image denoising on the to-be-processed magnetic resonance image by applying sparse dictionary learning, and performing Rician correction and multi-scale decomposition detail enhancement on the denoised image to obtain a detail-enhanced denoised image. According to the method, the imagedenoising effect contrast