论文部分内容阅读
针对光学相干层析(OCT)技术受散斑噪声影响严重的问题,研究了低噪声OCT成像技术。在系统中加入标定臂,利用标定信号对纵向扫描位置进行精确定位,降低检测过程中样品微位移所引入的误差;并在此基础上对各次纵向扫描信号进行曲波(curvelet)变换,采用自适应阈值法滤除图像散斑噪声,最后通过对多次纵向扫描信号的curvelet系数进行权重滤波和均值处理进一步降低散斑噪声。实验结果证明,与未经curvelet滤波的多次纵向扫描平均图像相比,低噪声OCT成像技术可以提高信噪比18.35%。
Aiming at the serious problem of speckle noise affected by optical coherence tomography (OCT), the low-noise OCT imaging technique is studied. The calibration arm is added in the system to locate the vertical scanning position accurately with the calibration signal to reduce the errors introduced by the micro-displacement of the sample in the detection process. Based on this, curvelet transformation is performed on each vertical scanning signal, The speckle noise is filtered by the thresholding method. Finally, the speckle noise is further reduced by weighting and averaging the curvelet coefficients of multiple longitudinal scanning signals. Experimental results show that low noise OCT imaging can improve the signal to noise ratio by 18.35% compared with the multiple vertical scanning averaging without curvelet filtering.