基于谱域光学相干层析系统中光谱涨落的指纹获取

摘要

Fingerprint plays a very important role in a criminal case as ' First Evidence' as a result of its specificity and stability. A variety of physical, chemical and optical techniques are available for the enhancement and detection of latent fingerprints. However, these existing techniques show some disadvantages such as harm to fingerprints, potential side effects, remaining trails and so on. Phase sensitive spectral domain optical coherence tomography ( SD-OCT ) for latent fingerprints detection has the advantages of non-contact, non-destructive, high-speed and high-sensitivity, but the results are greatly affected by the surface profile. In case of surface with large fluctuations, it is hard to extract the fingerprint with high contrast. In this paper, a method based on spectral fluctuation from spectral domain optical coherence tomo- graphy is proposed to acquire fingerprints. The proposed method only analyses and deals with the spectral fluctuation without retrieval of phase and depth, hence is simple, fast and unrelated to surface profile. The experiment results demonstrate that using the proposed method satisfied results of fingerprints can be obtained even on undulant surface.%指纹因其特异性和稳定性等特点而被称为"证据之首",在案件侦破中起着极其重要的作用.多种物理学、化学和光学技术都可以用于获取现场遗留的指纹,然而这些方法存在一些缺点,如会对指纹造成破坏、潜在的毒副作用、在现场留有痕迹等.利用谱域光学相干层析(spectral domain optical coherence tomography,简记为SD-OCT)技术进行指纹探测具有非接触、对指纹无损伤和高灵敏度的优势,利用OCT系统的相位敏感性我们可以在低对比条件下再现遗留在物体表面的指纹,但处理结果受指纹所在表面高低起伏影响,使得指纹信息对比度降低,难以被分离和提取.本文提出了一种基于干涉光谱涨落的指纹获取方法,只需对OCT系统得到的干涉光谱的涨落进行处理分析,即可得到遗留在物体表面的指纹图案,不需进行相位和表面轮廓高度的求取,算法简单、处理速度快,处理结果不受样品表面高低起伏的影响.实验结果表明,在起伏表面上,使用该方法也能较好地显现遗留在物体表面的指纹图案.