CAJ | 학술논문

本文针对频域光学相干层析成像技术(FDOCT)受到复共轭镜像影响严重的问题,提出一种K域抽取的色散编码稀疏重构算法(DEFR).通过在系统的参考臂和样品臂引入一定的色散量,利用数值色散补偿算法精确提取色散因子,采用稀疏重构算法对纵向扫描信号进行迭代提取峰值点及其镜像的频域光谱,减少了迭代中傅里叶变换的次数,实现了对复共轭镜像的压缩.实验结果表明,该算法在有效抑制镜像的同时运算效率提高了46.1%.
본문침대빈역광학상간층석성상기술(FDOCT)수도복공액경상영향엄중적문제,제출일충K역추취적색산편마희소중구산법(DEFR).통과재계통적삼고비화양품비인입일정적색산량,이용수치색산보상산법정학제취색산인자,채용희소중구산법대종향소묘신호진행질대제취봉치점급기경상적빈역광보,감소료질대중부리협변환적차수,실현료대복공액경상적압축.실험결과표명,해산법재유효억제경상적동시운산효솔제고료46.1%.
This paper describes the dispersion encoded sparse signal reconstruction algorithm based on k-domain extraction to suppress the complex conjugate mirror term in frequency domain Optical Coherence Tomography (OCT). This algorithm uses the numerical dispersion compensation to acquire the unmatched dispersion phase, meanwhile adopts the sparse signal reconstruction algorithm to extract the frequency spectrum of the peak and its corresponding mirror iteratively. This algorithm reduces the numbers of the FFT per iteration and cancels the complex conjugate mirror term. The experiment result demonstrated that the algorithm can suppress the complex conjugates effectively, and computational velocity improves 46.1%.