叠层衍射成像对比度反转校正的相位偏移方法
Phase offset method of ptychographic contrast reversal correction
AI总结:
针对直接叠层衍射成像中的对比度反转问题,提出相位偏移方法,通过操纵相位渐近线实现校正,优于离焦校正,适用于薄厚样品,并改善光学切片可解释性。
摘要:
直接叠层衍射成像方法(如单边带(SSB)方法)的对比度传递函数是单符号的,但这些方法有时仍会表现出对比度反转,最常见于投影势较强的地方。在较厚的样品中,中心聚焦通常能提供最佳的叠层衍射成像对比度,因为这会导致样品内部的离焦变化相互抵消。然而,远离入射表面聚焦通常是不理想的,因为这会使环形暗场(ADF)信号退化。在此,我们讨论了SSB叠层衍射成像频率响应中的相位包裹渐近线如何引起对比度反转,而无需动力学散射,以及如何通过操纵相位使渐近线移至更高频率或通过幅度调制进行阻尼来抵消这些反转。这正是实现采集后离焦校正对比度反转的关键。然而,我们在此引入的用于抵消对比度反转的相位偏移方法通常被发现优于采集后应用离焦的方法,具有更高的可靠性和通常更强的对比度。重要的是,相位偏移方法也适用于中心聚焦不适用的薄样品和厚样品。最后,该方法对焦点的独立性对于涉及叠层衍射成像的光学切片非常有用,通过更好地分离强势和焦点的影响来提高可解释性。
英文摘要:
The contrast transfer function of direct ptychography methods such as the single side band (SSB) method are single signed, yet these methods still sometimes exhibit contrast reversals, most often where the projected potentials are strong. In thicker samples central focusing often provides the best ptychographic contrast as this leads to defocus variations within the sample canceling out. However focusing away from the entrance surface is often undesirable as this degrades the annular dark field (ADF) signal. Here we discuss how phase wrap asymptotes in the frequency response of SSB ptychography give rise to contrast reversals, without the need for dynamical scattering, and how these can be counteracted by manipulating the phases such that the asymptotes are either shifted to higher frequencies or damped via amplitude modulation. This is what enables post collection defocus correction of contrast reversals. However, the phase offset method of counteracting contrast reversals we introduce here is generally found to be superior to post collection application of defocus, with greater reliability and generally stronger contrast. Importantly, the phase offset method also works for thin and thick samples where central focusing does not. Finally, the independence of the method from focus is useful for optical sectioning involving ptychography, improving interpretability by better disentangling the effects of strong potentials and focus.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。