Conference Agenda

Lensless imaging offers a compact and powerful route toward high-throughput quantitative optical metrology by replacing conventional image-forming optics with computational reconstruction. In this talk, I will discuss recent developments in lensless computational microscopy for quantitative phase and amplitude imaging, with emphasis on robust reconstruction strategies, improved measurement fidelity, and practical implementation aspects also within exotic spectral ranges (DUV nad NIR). Particular attention will be given to how lensless systems can enable high-throughput, compact, and cost-effective quantitative optical imaging while addressing challenges such as limited resolution, noise, and tomography. The presented concepts highlight the potential of lensless computational imaging as a flexible platform for modern optical metrology.

White-light interferometry (WLI) is a powerful and well-established optical measurement technique for the topographical characterization of surfaces. Due to its high precision, with height resolutions in the single-digit nanometer range, it is used in applications such as optical quality assurance. The technology relies on mechanical scanning of complex Mirau objectives to generate images and interference. As a result, such systems are large and expensive, sensitive to vibrations, and require a long measurement time.

This post presents a novel method for lensless WLI based on digital holography. The lensless design enables compact, lightweight measurement systems that require fewer individual images, thereby significantly reducing measurement time and sensitivity to vibrations. Additionally, larger fields of view can be captured or transmitted at the same resolution compared to lens-based systems.

The most challenging part of the lensless design is the illumination. While the conventional lens-based design relies on widely available broadband light sources, digital holography uses lasers, which are inherently monochromatic, due to the requirement of spatial and temporal coherence. The need for a multi-spectral illumination results in a unique set of requirements, which can be solved using supercontinuum lasers.

Digital Holography (DH) enables high-precision quantitative phase imaging but is often limited by coherent artifacts, such as parasitic multiple reflections. While these artifacts can be mitigated by averaging over several uncorrelated speckle illuminations, this approach is constrained by a finite axial correlation length. Beyond this length, signals decorrelate and are suppressed, posing a significant challenge for axially extended samples. Consequently, the object and calibration fields must be precisely depth-matched to prevent the loss of critical physical information. In this work, we demonstrate a method for retrieving accurate phase information of extended objects by numerically propagating both the illuminating speckle field and the object field. This approach maintains signal correlation across the entire object volume, effectively suppressing coherent artifacts while preserving the quantitative phase integrity of the extended sample.

We demonstrate an ultrasensitive interferometric technique that turns the normally unfavorable dark fringe into a highly sensitive phase magnifier. Using holographic readout in a slightly unbalanced Mach-Zehnder interferometer, we reconstruct the complex dark-port field by heterodyne time-holography and achieve sub-nanometer optical-path sensitivity within short acquisition times. Crucially, we extend the approach to quantitative phase imaging. For sufficiently small phase excursions, or optical thickness variations, the amplified phase image remains linear and enables direct quantitative reconstruction with enhanced sensitivity. For larger excursions, the nonlinear response yields highly accurate phase-contour maps, revealing fine phase structure with exceptional precision. This dual capability establishes a compact and versatile platform for ultrasensitive phase metrology and opens a realistic route toward ultraprecise quantitative phase imaging of biological samples.

In interferometric phase retrieval, the sign of a reconstructed wavefront, phase map, or height map is often inferred from setup-dependent cues. Vibration can corrupt these cues and produce a globally inverted result. We present a stored-reference method for detecting and correcting such inversions, illustrated for digital holography. Each new reconstruction is compared with a sample-specific stored reference using orientation-sensitive observables for a flip/no-flip decision. Representative vibration-affected measurements show that the method restores the expected sign relative to the reference and improves the reliability of downstream phase evaluation.