A device and method for forming an image of a sample includes illuminating the sample with a light source; acquiring a plurality of images of the sample using an image sensor, the sample being placed between the light source and the image sensor, no magnifying optics being placed between the sample and the image sensor, the image sensor lying in a detection plane, the image sensor being moved with respect to the sample between two respective acquisitions, such that each acquired image is respectively associated with a position of the image sensor in the detection plane, each position being different from the next; and forming an image, called the high-resolution image, from the images thus acquired.

A method for defect inspection of a transparent substrate comprises utilizing a wavefront reconstruction unit to obtain complex defect diffraction wavefront of a transparent substrate; using a complex defect diffraction module to confirm the effective diffraction distance of the complex defect diffraction wavefront; utilizing a defect detection module to detect position of the defect of the transparent substrate; using a defect classification module to perform extraction, analysis and classification of diffraction characteristics and utilizing a machine learning algorithm or a deep learning algorithm to automatically identify the defects.

A method of performing phase retrieval and holographic image reconstruction of an imaged sample includes obtaining a single hologram intensity image of the sample using an imaging device. The single hologram intensity image is back-propagated to generate a real input image and an imaginary input image of the sample with image processing software, wherein the real input image and the imaginary input image contain twin-image and/or interference-related artifacts. A trained deep neural network is provided that is executed by the image processing software using one or more processors and configured to receive the real input image and the imaginary input image of the sample and generate an output real image and an output imaginary image in which the twin-image and/or interference-related artifacts are substantially suppressed or eliminated. In some embodiments, the trained deep neural network simultaneously achieves phase-recovery and auto-focusing significantly extending the DOF of holographic image reconstruction.

A method of directing a wavefront of coherent radiation through a sample of objects in a suspension, capturing an interference pattern between the wavefront of coherent radiation and a wavefront of the diffracted by the object with an image sensor, numerically determining the focal plane of at least one object, and numerically reconstructing a de-focused image of the at least one object from the interference pattern in an image plane which is substantially parallel to the image sensor and in a plane with a predetermined offset from the focal plane. The method further includes identifying at least one portion in the defocused image corresponding to the at least one object in the sample, and calculating from each of said portions at least one feature of the corresponding object.

A method of extracting particles from a two-dimensional (2D) hologram recorded as part of a digital inline holography system includes reconstructing a three-dimensional (3D) optical field from the recorded 2D hologram. In addition, particles are extracted/segmented from the 3D optical field, wherein segmented particles are identified by particle location in three-dimensional space and a cross-sectional area of the segmented particle. Based on the identified particle location and cross-sectional area, extracted particles are removed from the 2D hologram to generate an updated 2D hologram. These steps are repeated iteratively until a threshold is met.

A method of correcting a holographic image, a processing device, a dark field digital holographic microscope, a metrology apparatus and an inspection apparatus. The method includes obtaining a holographic image; determining at least one attenuation function due to motion blur from the holographic image; and correcting the holographic image, or a portion thereof, using the at least one attenuation function.

A method for estimating a depth map associated with a hologram representing a scene, the method includes steps of: reconstruction of images of the scene, each image being associated with a depth; decomposition of each image into a plurality of thumbnails adjacent to each other, each thumbnail being associated with the depth and including a plurality of pixels; determination, for each thumbnail, of a focus map by supplying, at the input of a network of neurons, values associated with the pixels of the thumbnail, to obtain, at the output of the network, the focus map including a focus level associated with the pixel concerned; and determination of a depth value, for each point of a depth map, as a function of the focus levels obtained. The invention also relates to an estimation device and an associated computer program.

Digital holographic microscopy and related image processing techniques are described. A hologram captured in an image frame is split into different depths while a new hologram is being captured. Image slices of the hologram are determined and using free space impulse responses that are pre-calculated at a different precision than processing operations using the holographic data. Each computation is calculated in parallel based on the number of available processing cores and threads. The image slices are combined into a 2D array or 3D array to permit further processing of the combined array to count and size particles in the image frame. The reconstructed hologram is displayed at a subsequent image frame than that used to capture the hologram.

The invention relates to a device, such as a digital holographic microscope (1), for detecting and processing a first full image of a measurement object, measured with a first offset, wherein an arrangement is provided for generating at least one further full image with at least one offset that differs from said first offset.

In the present invention, by receiving three-dimensional spatial data for a three-dimensional object, calculating a size of an elemental fringe pattern for each depth and a maximum size of the elemental fringe pattern for the three-dimensional spatial data, dividing the entire resolution of an entire predetermined hologram into a plurality of segments based on the number of divisions of the horizontal direction and the vertical direction, expanding resolution for each segment based on a maximum size for the elemental fringe pattern size, calculating a plurality of segment unit holograms for each of the segments, and accumulating overlapping regions between the plurality of segment unit holograms to produce the entire hologram for the entire resolution, it is possible to provide a hologram generating apparatus and method which may generate a hologram having a higher resolution at a faster speed than the conventional one.