论文部分内容阅读
为了提高光声成像(PAT)的对比度和分辨率,需对组织样品的光声(PA)信号进行基于探头脉冲响应的滤波反卷积以恢复其频谱特性。对宽带光声信号而言,由于带通滤波的截止频率由人为确定,噪声不能得到有效抑制,很难获得稳定的反卷积结果。针对此问题,提出了基于维纳滤波反卷积的光声成像方法,利用点光声源获得超声探头的脉冲响应。利用维纳滤波抑制反卷积过程中噪声的影响,滤波器参数由离散小波变换(DWT)动态估计,样品光声图像由时域后向投影算法重建。数值模拟与成像实验均表明该方法有效地抑制了噪声对反卷积的影响,提高了光声成像的对比度和分辨率。
In order to increase the contrast and resolution of photoacoustic imaging (PAT), a probe deconvolution based on the probe impulse response needs to be performed on the photoacoustic (PA) signal of the tissue sample to restore its spectral characteristics. For broadband photoacoustic signals, since the cut-off frequency of the band-pass filter is determined by man and the noise can not be effectively suppressed, it is difficult to obtain a stable deconvolution result. To solve this problem, a photoacoustic imaging method based on Wiener filter deconvolution is proposed, and the impulse response of the ultrasonic probe is obtained by using the point light source. Wiener filter was used to suppress the influence of noise in the deconvolution process. The filter parameters were estimated dynamically by discrete wavelet transform (DWT), and the photoacoustic images were reconstructed by the time-domain backward projection algorithm. Both numerical simulation and imaging experiments show that this method can effectively reduce the effect of noise on deconvolution and improve the contrast and resolution of photoacoustic imaging.