Hypercentric Imaging in Coherent Optical Metrology (HyperCOMet)

在生产过程中，技术部件会受到不必要的几何变形和表面缺陷的影响，如断裂、划痕和凹痕。因此，板形检验是工业质量保证的必备条件。如果需要高精度，干涉测量方法是很成熟的，因为它们是非接触式的，测量不确定度可以精确到照明波长的一小部分。微冷成形和微系统工程的进步允许生产越来越复杂的形状微型零件，这给光学计量学带来了新的挑战。因此，除了高精度，未来的干涉测量系统必须通过提供在合理的时间范围内测量高度复杂的形状物体的能力来满足这些要求。目前基于标准成像系统的现有系统无法完成这一任务，因为它们的视角有限，而且测量复杂几何图形的完整3D形式通常需要从不同观察方向进行多次单独测量。这需要大量的硬件，并涉及耗时的校准和拼接过程来重建对象的形状。解决这个问题的方法可能是超中心(或中心)成像，这是最近被引入机器视觉领域的。它使凸起物体的周向可见性(正面和侧面的同时成像)成为可能。目前，该技术主要与图像处理结合使用，例如用于快速检测物体表面纹理的不规则性。虽然干涉面形测量技术可以很大程度上受益于这种成像方式，但据我们所知，超中心成像在相干光学计量领域的应用尚未见报道。本项目旨在将超中心成像引入干涉面形测量领域。期望这一概念在复杂形状物体的工业质量检测中为物体具有凸形的所有情况提供快速和精确的形状确定。示例性地，将采用多波长相移干涉(MW-PSI)和频域白光干涉(FD-WLI)这两种相干光学形状测量方法。这需要在波动光学水平上理解超中心成像过程，重点关注图像平面中光的散斑统计和相干性。此外，必须开发特定的几何模型，该模型将观察到的路径差异与对象的实际形状联系起来，同时考虑到超中心透镜的反向放大。该项目的结果可能导致新一代紧凑、灵活和精确的干涉传感器，提供测量不确定度精确到亚微米范围的周向形状数据。

During production, technical parts are subject to unwanted geometrical distortions and surface defects such as fractures, scratches and dents. Therefore, shape inspection is a mandatory perquisite for industrial quality assurance. If high precision is required, interferometric methods are well established, because they work contactless and provide a measurement uncertainty down to a fraction of the illumination’s wavelength.Advancements in micro cold forming and micro systems engineering allow for the production of increasingly complex shaped micro parts, which leads to new challenges in optical metrology. Thus, in addition to the high precision, future interferometric measurement systems have to meet these requirements by offering the ability to measure highly complex shaped objects in a reasonable time frame. Currently available systems based on standard imaging systems cannot accomplish this task, because they have a limited viewing angle, and measuring the full 3D form of a complex geometry often requires a number of individual measurements from different observation directions. This requires extensive hardware and involves time-consuming calibration and stitching procedures to reconstruct the object’s form.A solution to this problem could be hypercentric (or pericentric) imaging, which has recently been introduced to the field of machine vision. It enables circumferential visibility (simultaneous imaging of the front and the side surfaces) of convex objects. Currently the technique is mainly used in combination with image processing, e.g. to rapidly detect irregularities of an object's surface texture. Even though interferometric shape measurement techniques could largely benefit from this imaging modality, an application of hypercentric imaging to the field of coherent optical metrology has, to the best of our knowledge, not been reported.The aim of the proposed project is to introduce hypercentric imaging to the field of interferometric shape measurement. It is expected that this concept provides fast and precise form determination in industrial quality inspection of complex shaped objects for all cases in which the objects have convex shape. Exemplarily, the two coherent optical shape measurement methods of multi wavelengths phase shifting interferometry (MW-PSI) and frequency domain white light interferometry (FD-WLI) will be adapted. This requires understanding of the hypercentric imaging process on a wave optics level with a focus on speckle statistics and coherence properties of the light in the image plane. Furthermore, a specific geometry model has to be developed, that links the observed path differences to the actual shape of the object, while taking the reversed magnification of hypercentric lenses into account. The results of the project could lead to a new generation of compact, flexible and precise interferometric sensors offering circumferential shape data with a measurement uncertainty down to the sub micrometer range.