基于特征值分解的时间反转超声成像技术

Time-reversal ultrasound imaging technology based on eigenvalue decomposition

由于声波衍射的关系,超声成像系统的检测效果会受到检测深度、传播介质等因素的影响,难以取得理想的高分辨率成像结果。文中基于时间反转声学理论,将基于时间反转矩阵特征值分解的成像算法引入超声成像技术中。在该算法中,每一个非零的特征值均对应一个散射体,通过对时间反转矩阵进行特征值分解得到包含检散射体相位信息的特征向量,将得到的时间反转信号反向发射从而实现对不同散射体的成像。仿真实验结果表明,该算法对于一定深度的缺陷具有更好的检测效果,且缺陷的横向分辨率明显优于一般超声成像算法,这一结果有益于提升超声系统区分多个近距离目标的能力。

Influenced by factors such as acoustic wave diffraction,the detection effect of ultrasound imaging system will be affected by the detection depth,and it is difficult to obtain ideal imaging results. Based on the time-reversal acoustic theory,an imaging algorithm based on eigenvalue decomposition of time reversal matrix is introduced into ultrasonic imaging technology. In proposed algorithm,each non-zero eigenvalue corresponds to a scatterer,and the eigenvalue decomposition of time reversal matrix is performed to obtain a eigenvector containing phase information of scatterers are obtained,The imaging of different scatterers can be realized by transmitting the time reversal signal backward. The simulation results show that this algorithm has better detection effect for certain depth defects,and the lateral resolution of defects is obviously better than that of general ultrasound imaging algorithm. This result is beneficial to enhance the ability of ultrasound system to distinguish multiple short-range targets.