基于非均匀快速傅里叶变换的干涉合成孔径显微算法

Interferometric Synthetic Aperture Microscopy Algorithm Based on Nonuniform Fast Fourier Transform

在频域光学相干层析(SDOCT)系统中,样品仅在焦深范围内有高横向分辨率,在焦外分辨率降低。干涉合成孔径显微术(ISAM)是一种三维图像重构算法,可以改善离焦区的图像模糊状况,达到在所有成像深度中都可以获得焦平面处横向分辨率的效果,同时可以解决SDOCT中系统的横向分辨率和成像深度之间的矛盾。介绍了ISAM重构算法的原理和适用的范围,对比分析了传统SDOCT成像算法和经过ISAM重构算法处理的成像图像结果。将非均匀快速傅里叶变换引入ISAM算法,实验结果表明该方法极大地节省了运算时间,提升了实时成像高分辨率SDOCT系统的性能。

In the spectral domain optical coherence tomography （SDOCT） system, we can get high transverse resolution only within the focal depth field. The transverse resolution degrades away from the focal plane. Interferometric synthetic aperture microscopy （ISAM） is a three-dimensional image reconstruction algorithm, which can make the out-of-focus image clearer, and a resolution in all planes that is equivalent to the resolution at the focal plane can be obtained. At the same time, the contradiction between the transverse resolution and the imaging depth in the SDOCT system can be solved. The principle of ISAM and its possible applications are introduced and the comparison between the reconstructed images by the traditional SDOCT method and the ISAM is performed. The nonuniform fast Fourier transform （NUFFT） algorithm is applied in ISAM. Experimental results show that the NUFFT can greatly reduce the operation time, improving the performance of real-time and high-resolution SDOCT system.