特征空间最小方差波束形成与相关系数特征值加权相融合的超声成像算法  

Ultrasound imaging algorithm by combining eigenspace-based minimum variance beamforming and correlation coefficient eigenvalue weighting

作　　者：郑驰超[1] 彭虎[1] 

机构地区：[1]合肥工业大学生物医学工程系,合肥230009

出　　处：《声学学报》2016年第1期25-29,共5页Acta Acustica

基　　金：国家自然科学基金青年科学基金(61201060);国家自然科学基金(61172037)资助

摘　　要：为了进一步提高超声成像的质量,提出一种信号特征空间的最小方差波束形成与相关系数特征值加权相融合的超声成像算法。利用超声回波信号具有一定的相关性,而相关系数空间最大特征值可以反映回波信号相关性较强的性质,将该特征值作为自适应加权系数对信号特征空间最小方差波束形成(EIBMV)的结果进行加权成像,得到高质量的成像结果。通过对散射点目标和斑目标的Field II仿真,结果表明该算法相对于EIBMV算法,亮斑对比度提高了4.22 dB,暗斑对比度提高了1.88 dB,并且进一步提高了横向分辨率。To improve the ultrasonic image quality, in this paper, we propose an imaging algorithm by combining eigenspace-based minimum variance beamforming and correlation coefficient eigenvalues weighting. Due to the facts that the ultrasonic echo signals are relatively related, and that the proposed maximum eigenvalue of correlation coefficient space is capable of reflecting the related intensity, the adaptive weighting coefficients based on the eigenvalues are employed for the imaging of eigenspace-based minimum variance adaptive （EIBMV） beamforming, leading to high-quality imaging. The Field II simulation results of both point targets and cyst phantom in a speckle pattern demonstrate that by using our proposed algorithm the contrast of massive cyst was improved by 4.22 dB, and meanwhile the contrast of anechoic cyst was enhanced by 1.88 dB in comparison to EIBMV algorithm. Besides, the lateral resolution was also improved.

关 键 词：最大特征值 成像算法 特征空间 加权系数 相关系数 波束形成 最小方差 超声成像 

分 类 号：O426.9[理学—声学]