Method and system for reconstruction of an image from electron diffraction data

Abstract

Provided here is a method for generating a reconstructed image (500) from electron diffraction data of a test sample (100), comprising: - receiving the electron diffraction data of the test sample (100) comprising a plurality of test event-detected counts (210a, 210b, 210c); - selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g. 310b, e, h) (selected kernel image (e.g. 310b, e, h)) from a kernel image library (300) comprising a plurality of kernel images (e.g. 310a to i); wherein each kernel image of the kernel image library (300) (e.g. 310a to i) is a pre-calculated mosaic for adding to a reconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500); and - generating the reconstructed image (500), comprising adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400) thereby gradually mosaicking and rendering the reconstructed image (500).

Description

[0001]METHOD and SYSTEM FOR RECONSTRUCTION OF AN IMAGE FROM ELECTRON DIFFRACTION DATA Field of the invention The present invention is in a field of image reconstruction from electron diffraction data. Background to the invention Recent trends in the field of scanning transmission electron microscopy (STEM) have led to the implementation of powerful methods for the direct retrieval of the electrostaticpotential of a solid-state specimen. Those include the integration of the centre of mass ofthe far-field intensity (iCoM) and ptychography, which was first pioneered in the field of X-ray microscopy. Electron ptychography exists in different forms, including iterative reconstruction schemes and analytical approaches. Analytical ptychography itself includes side-band integration (SBI), alternatively known as single side-band (SSB) reconstruction, and Wigner distribution deconvolution (WDD). An interest of those potential retrieval methods is their inherent dose-efficiency. STEM experiments are often associated to a strong probability of beam damage, which makes the imaging of dose-sensitive materials difficult. In that context, performing a measurement with the most efficient use of each detected electrons is crucial to extending known experimental approaches to new scientific fields. One of such fields is structural biology, where less sophisticated TEM approaches are now common, and whose primary experimental limitation is the high likelihood of beam damage. In that context, any breakthrough leading to an increment in recovered information per invested electrons may lead to major advancements. Typically, both iCoM and ptychography require the prior recording of a collection of diffraction patterns, corresponding to the lateral scan raster of a convergent electron probe. This experimental set-up is often referred to as 4D-STEM, as it leads to aresolution both in real- and reciprocal space, hence in four recording dimensions. One ofthe main challenges of such experiments is the readout speed of the detectors used. Even in 1 bit mode, frame times are at best 80 µs for Medipix3 type detectors. As a result, 4D-STEM is often time-intensive, which makes it prone to specimen drift and beam damage. This is in part why it is important to introduce faster options for both data acquisition and treatment, even going toward live imaging, as has been explored by other groups. A second challenge is the amount of data generated as part of the experiment, which can reach dozens of GB in just a few seconds. Transferring this data for further processing implies burdensome needs in terms of data bandwidth and computational power, resulting in hardware that is very expensive, specialistic and difficult to maintain. The present invention aims at providing a new framework for performing analyticalptychography, here including the SBI and WDD approaches but also extended to theiCoM method, with a minimized cost in invested electrons, shorter acquisition times and a lower need in data storage and transfer. For this purpose, new developments in event- driven electron detection are exploited. Summary of the inventionProvided herein is a method for generating a reconstructed image (500) from electrondiffraction data of a test sample (100), comprising:- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) from a kernel image library (300) comprising a plurality of kernel images (e.g. 310b, e, h), wherein: -a kernel image (e.g.310b, e, h) is pre-calculated and is a mosaic for adding to areconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500); and- generating the reconstructed image (500), comprising:- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) thereby gradually mosaicking and rendering the reconstructed image (500).Provided herein is a method for generating a reconstructed image (500) from electrondiffraction data of a test sample (100), comprising:- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) - a selected kernel image (310b, 310e, 310h) - from a kernel image library(300) comprising a plurality of kernel images (310a to i), wherein:- each kernel image (e.g.310a to i) of the kernel image library (300) is pre-calculated and is a mosaic for adding to a reconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500); and- generating the reconstructed image (500), comprising:- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) thereby gradually mosaicking and rendering the reconstructed image (500). According to a preferred aspect, each kernel image (310a to i) of the kernel image library (300): -has a size that is a portion of the size of the reconstruction grid (400);- is simulated from a single simulated event-detected count having a simulatedcount grid vector location (^^⃗^) on a simulated count grid of a simulated countdetector. According to a preferred aspect:- the electron diffraction data further comprises a plurality of different scan grid vectorlocations (100a, 100b, 100c, ^^⃗), each scan grid vector location being a location of a test electron probe that interacts with the test sample (100) at vector position (^^⃗) of the scan grid (120) thereby generating an exit wave (120a) from the test sample (100);- each test event-detected count (210a, 210b, 210c) of the plurality of test event-detectedcounts (210a, 210b, 210c) has a test count grid vector location (^^⃗^) on a count grid (205)of the event-driven detector (200);- the test count grid vector location (^^⃗^) is linked in the electron diffraction data to thescan grid vector location (100a, 100b, 100c, ^^⃗) that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) at the test count grid vector location (^^⃗^);- the kernel image (310b, 310e, 310h) is selected when (^^⃗^) matches (^^⃗^);- the adding of each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) is at a collection of reconstruction grid vector locations (^⃗) centered on the scan grid vector location (^^⃗) that generated the exit wave (120a) causing the test event- detected count (210a, 210b, 210c) on which the kernel image (310b, 310e, 310h) was selected.Also provided herein is a method for generation of a reconstructed image (500) fromelectron diffraction data of a test sample (100), comprising:- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c), wherein the electron diffraction data comprises: -a plurality of different scan grid vector locations (100a, 100b, 100c, ^^⃗), eachscan grid vector location being a location of a test electron probe that interacts with the test sample (100) at vector position (^^⃗) of the scan grid (120) thereby generating an exit wave (120a) from the test sample (100); -for each scan grid vector location (^^⃗), at least one test event-detected count(210a, 210b, 210c) caused by the exit wave (120a) interacting with an event- driven detector (200), wherein each test event-detected count (210a, 210b, 210c) has a test count grid vector location (^^⃗^) on a count grid (205) of the event-driven detector (200);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) from a kernel image library (300), wherein: -the kernel image library (300) comprises a plurality of pre-calculated kernelimages (310a to i), wherein: -each kernel image (310a to i) of the plurality:- is a mosaic for adding to a reconstruction grid (400) of thereconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500); -has a size that is a portion of the size of the reconstruction grid(400); -is simulated from a single simulated event-detected count havinga simulated count grid vector location (^^⃗^) on a simulated countgrid of a simulated count detector; -the kernel image (310b, 310e, 310h) is selected when (^^⃗^) matches (^^⃗^);- generating the reconstructed image (500), comprising:- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) to a collection of reconstruction grid vector locations (^⃗) centered aroundthe scan grid vector location (^^⃗) that generated the test event-detected count (210a, 210b, 210c) (on which the kernel image (310b, 310e, 310h) was selected),thereby gradually mosaicking and rendering the reconstructed image (500). According to a preferred aspect the adding of each kernel image (310b, 310e, 310h) to the reconstruction grid (400) comprises a step of summing individual pixel intensities of the added kernel image (310b, 310e, 310h) at each reconstruction grid vector location (^⃗) with previous individual pixel intensities at the same reconstruction grid vector locations (^⃗). According to a preferred aspect the adding of each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400) at a reconstruction grid vector location (^⃗) is adjacent, overlapping, or superimposing, wherein: -the adjacent adding comprises positioning the selected kernel image (310b,310e, 310h) next to an existing kernel image on the reconstruction grid (400); -the overlapping adding comprises positioning the selected kernel image (310b,310e, 310h) so that one or more of its edges overlaps with a part of one or more existing kernel images on the reconstruction grid (400); -the superimposing adding comprises positioning the selected kernel image(310b, 310e, 310h) exactly over one or more existing kernel images on the reconstruction grid (400).The existing kernel image on the reconstruction grid (400) is a selected kernel imageadded previously to the reconstruction grid (400) during the gradual mosaicking andrendering of the reconstructed image (500).According to a preferred aspect after the selected kernel image (310b, 310e, 310h) is added to the reconstruction grid (400), the reconstruction grid becomes updated, and only a portion of the reconstruction grid (400) where the (selected) kernel image (310b, 310e, 310h) has been added becomes updated. According to a preferred aspect the size of each kernel image (310a to i) of the kernel image library (300) is equal to or less than 50% of the size of the reconstruction grid (400) by area. According to a preferred aspect the simulation of each kernel image (310a to i) in the image library (300) from a single simulated count having a simulated count grid vectorlocation (^^⃗^) comprises analytical ptychography.According to a preferred aspect:- the analytical ptychography is integrated centre of mass (iCOM), side-bandintegration (SBI), or Wigner distribution deconvolution (WDD); and -(the simulation of) the kernel image (310a to i) (in the image library (300))represents a partial measurement of an electrostatic potential (μ) map weighted by an interaction constant (σ), of a charge density (ρ) map weighted by the same interaction constant (σ), or of the real or imaginary parts of a transmission function (T) map, wherein transmission function is the complex exponential of the electrostatic potential (μ) weighted by the interaction constant (σ). According to a preferred aspect the generating of the reconstructed image (500) is performed in real time.Further provided is a method for interactively generating a reconstructed image (500)from electron diffraction data of a test sample (100), comprising the method asdescribed herein, wherein: - the method further comprises sending an instruction to an electron microscope (EM)to direct a test electron probe to a scan grid vector location (100a, 100b, 100c, ^^⃗), suchthat the electron probe interacts with the test sample (100) and generates an exit wave(120a);- in the step of:receiving the electron diffraction data of the test sample (100) comprising a plurality of test event-detected counts (210a, 210b, 210c) caused by the generated exit wave (120a),said electron diffraction data of the test sample (100) is generated by the electronmicroscope (EM);- in the step of:adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400) thereby gradually mosaicking and rendering the reconstructed image (500), each selected kernel image (310b, 310e, 310h) is added to the reconstruction grid (400)at a collection of reconstruction grid vector locations (^⃗) centered on the scan grid vectorlocation (^^⃗) that generated the exit wave (120a) causing the test event-detected count(210a, 210b, 210c) on which the kernel image (310b, 310e, 310h) was selected;- the method further comprises: repeating the sending, receiving, selecting and adding, wherein for at least one repetition, a different scan grid vector location (100a, 100b, 100c, ^^⃗) is used,thereby gradually mosaicking and rendering the reconstructed image (500).Further provided is a (interactive) method for interactively generating a reconstructedimage (500) from electron diffraction data of a test sample (100), comprising:- sending an instruction to an electron microscope (EM) to direct a test electron probe toa scan grid vector location (100a, 100b, 100c, ^^⃗), such that the electron probe interacts with the test sample (100) and generates an exit wave (120a);- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c) caused by the generated exit wave (120a);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) from a kernel image library (300) comprising a plurality of kernel images (310a to i), wherein: -each kernel image (e.g.310a to i) of the kernel image library (300) is a pre-calculated mosaic for adding to a reconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500);- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400)at a collection of reconstruction grid vector locations (^⃗) centered on the scan grid vector location (^^⃗) that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) on which the kernel image (310b, 310e, 310h) was selected,- repeating the sending, receiving, selecting and adding, wherein for at least onerepetition, a different scan grid vector location (100a, 100b, 100c, ^^⃗) is used, thereby gradually mosaicking and rendering the reconstructed image (500).According to a preferred aspect the (interactive) method according incorporates the anyaspect of the additional subject matter of the method for generating a reconstructed image (500) from electron diffraction data of a test sample (100) as described herein.The presently described method is a computer-implemented method.Further provided is a computing device or system configured for performing the(computer implemented) method according as described herein.Further provided is a computer program or computer program product having instructionswhich when executed by a computing device or system cause the computing device or system to perform the (computer implemented) method as described herein.Further provided is a computer readable medium having stored thereon a computerprogram having instructions which when executed by a computing device or system cause the computing device or system to perform the (computer implemented) method as described herein. Figure LegendsFIG.1 is a flow diagram illustrative of the methods and systems described herein.Detailed description of invention Before the present system and method of the invention are described, it is to be understood that this invention is not limited to particular systems and methods or combinations described, since such systems and methods and combinations may, of course, vary. It is also to be understood that the terminology used herein is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims. As used herein, the singular forms "a", "an", and "the" include both singular and plural referents unless the context clearly dictates otherwise. The terms "comprising", "comprises" and "comprised of" as used herein are synonymous with "including", "includes" or "containing", "contains", and are inclusive or open-ended and do not exclude additional, non-recited members, elements or method steps. It will be appreciated that the terms "comprising", "comprises" and "comprised of" as used herein comprise the terms "consisting of", "consists" and "consists of". The recitation of numerical ranges by endpoints includes all numbers and fractions subsumed within the respective ranges, as well as the recited endpoints. The term "about" or “approximately” as used herein when referring to a measurable value such as a parameter, an amount, a temporal duration, and the like, is meant to encompass variations of + / -10% or less, preferably + / -5% or less, more preferably + / -1% or less, and still more preferably + / -0.1% or less of and from the specified value, insofar such variations are appropriate to perform in the disclosed invention. It is to be understood that the value to which the modifier "about" or “approximately” refers is itself also specifically, and preferably, disclosed. Whereas the terms “one or more” or “at least one”, such as one or more or at least one member(s) of a group of members, is clear per se, by means of further exemplification, the term encompasses inter alia a reference to any one of said members, or to any two or more of said members, such as, e.g., any ≥3, ≥4, ≥5, ≥6 or ≥7 etc. of said members, and up to all said members. All references cited in the present specification are hereby incorporated by reference in their entirety. In particular, the teachings of all references herein specifically referred to are incorporated by reference. Unless otherwise defined, all terms used in disclosing the invention, including technical and scientific terms, have the meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. By means of further guidance, term definitions are included to better appreciate the teaching of the present invention. In the following passages, different aspects of the invention are defined in more detail. Each aspect so defined may be combined with any other aspect or aspects unless clearly indicated to the contrary. In particular, any feature indicated as being preferred or advantageous may be combined with any other feature or features indicated as being preferred or advantageous. Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to a person skilled in the art from this disclosure, in one or more embodiments. Furthermore, while some embodiments described herein include some but not all other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those in the art. For example, in the appended claims, any of the claimed embodiments can be used in any combination. In the present description of the invention, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration only of specific embodiments in which the invention may be practiced. Parenthesized or emboldened reference numerals affixed to respective elements merely exemplify the elements by way of example, with which it is not intended to limit the respective elements. Unless otherwise indicated, all figures and drawings in this document are not to scale and are chosen for the purpose of illustrating different embodiments of theinvention. In particular, the dimensions of the various components are depicted inillustrative terms only, and no relationship between the dimensions of the various components should be inferred from the drawings, unless so indicated. It is to be understood that other embodiments may be utilised and structural or logical changes may be made without departing from the scope of the present invention. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims. &&Core Provided herein is a method or system for generation of a reconstructed image (500) from electron diffraction data of a test sample (100). A flow diagram illustrative of themethod is presented in FIG. 1. The method or system receives electron diffraction dataof the test sample (100) collected by an event driven detector (200), and uses a pluralityof pre-calculated kernel images (310b, e, h) to gradually mosaic and render thereconstructed image (500) on a reconstruction grid (400), wherein each pre-calculatedkernel image (310 b, e, h) is selected from a kernel image library (300), the selectionbased on a count grid vector location (^^⃗^) of the detected event (210a, 210b, 210c) on acount grid (210) of the event driven detector (200). The vector locations (^⃗) of theselected kernel image (310 b, e, h) on the reconstruction grid (400) are determined froma scan grid vector location (^^⃗) that is a location where a test electron probe (105)interacts with the test sample (100) on a scan grid (120).Thus, according to one aspect, provided is a method for generating a reconstructedimage (500) from electron diffraction data of a test sample (100), comprising:- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) from a kernel image library (300) comprising a plurality of kernel images (e.g. 310a to i), wherein: -a kernel image (e.g.310a to i) (of the kernel image library (300)) is pre-calculated and is a mosaic for adding to a reconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500);- generating the reconstructed image (500), comprising:- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) thereby gradually mosaicking and rendering the reconstructed image (500). According to a preferred aspect:- each kernel image (310a to i) of the kernel image library (300):- has a size that is a portion of the size of the reconstruction grid (400);- is simulated from a single simulated event-detected count having a simulatedcount grid vector location (^^⃗^) on a simulated count grid of a simulated countdetector; According to a preferred aspect:- the electron diffraction data further comprises a plurality of different scan grid vectorlocations (100a, 100b, 100c, ^^⃗), each scan grid vector location being a location of a test electron probe that interacts with the test sample (100) at vector position (^^⃗) of the scan grid (120) thereby generating an exit wave (120a) from the test sample (100);- each test event-detected count (210a, 210b, 210c) of the plurality of test event-detectedcounts (210a, 210b, 210c) has a test count grid vector location (^^⃗^) on a count grid (205)of the event-driven detector (200);- the test count grid vector location (^^⃗^) is linked in the electron diffraction data to thescan grid vector location (100a, 100b, 100c, ^^⃗) that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) at the test count grid vector location (^^⃗^);- the kernel image (310b, 310e, 310h) is selected when (^^⃗^) matches (^^⃗^);- the adding of each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) is at a collection of reconstruction grid vector locations (^⃗) centered on the scangrid vector location (^^⃗) that generated the exit wave (120a) causing the test event- detected count (210a, 210b, 210c) on which the kernel image (310b, 310e, 310h) was selected.According to a more detailed aspect, provided is a method for generating a reconstructedimage (500) from electron diffraction data of a test sample (100), comprising:- receiving the electron diffraction data of the test sample (100), wherein the electrondiffraction data comprises: -a plurality of different scan grid vector locations (100a, 100b, 100c, ^^⃗), eachscan grid vector location being a location of a test electron probe that interacts with the test sample (100) at vector position (^^⃗) of the scan grid (120) thereby generating an exit wave (120a) from the test sample (100); -for each scan grid vector location (^^⃗), at least one test event-detected count(210a, 210b, 210c) caused by the exit wave (120a) interacting with an event- driven detector (200), wherein each test event-detected count (210a, 210b, 210c) has a test count grid vector location (^^⃗^) on a count grid (205) of the event-driven detector (200);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) from a kernel image library (300), wherein: -the kernel image library (300) comprises a plurality of pre-calculated kernelimages (310a to i), wherein: -each kernel image (310a to i) of the plurality:- is a mosaic for adding to a reconstruction grid (400) of thereconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500); -has a size that is a portion of the size of the reconstruction grid(400); -is simulated from a single simulated event-detected count havinga simulated count grid vector location (^^⃗^) on a simulated countgrid of a simulated count detector; -the kernel image (310b, 310e, 310h) is selected when (^^⃗^) matches (^^⃗^);- generating the reconstructed image (500), comprising:- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) at a collection of reconstruction grid vector locations (^⃗) centered on thescan grid vector location (^^⃗) that generated the test event-detected count (210a,210b, 210c) (on which the kernel image (310b, 310e, 310h) was selected),thereby gradually mosaicking and rendering the reconstructed image (500). In FIG.1, a portion of the reconstruction grid (400) is shown. The portion is represented as a dashed box on the reconstructed image (500). In a typical prior art method or system, diffraction patterns are recorded using a 2D fullframe imaging sensor that reads out a full frame 2D image of the diffraction pattern (2D-FF-DP, 2D full frame diffraction pattern). As result of obtaining multiple 2D-FF-DPs acquired at different scan grid vector locations (^^⃗), a reconstructed image can be determined. However, the determination is typically performed offline, using multiplestored 2D-FF-DPs, and requires a high-powered processor(s) to perform the necessaryburden of transforming and combining of each 2D-FF-DP into the reconstructed image.Typically, processing times range from less than an hour to a full day. Typical storagerequirements are high because of the full frame-based readout (e.g.2 to 32 GB).By creating a kernel image library (300) of pre-calculated kernel images (310 a to i), the reconstructed image can be created in real-time during acquisition, because most of the necessary calculation has been performed in advance. Thus, computation time is greatly reduced, and it becomes possible to obtain the final reconstructed image at the end of the acquisition. An advantage of such real-time reconstruction is an ability to re-direct the test electron probe (105) to regions of the scan grid (120) giving rise to poorer quality data (e.g. due to sample thickness), in order to improve it locally. The interactivity or reactivity henceallows for real-time adjustments. Those adjustments are impossible in the off-linescenario.The present method and system also permits a reduction in beam damage, sinceacquisition can be stopped once the reconstructed image is of sufficient quality, and / oronly local regions of the sample that need additional data are exposed to longer beamdwell times. By comparison, in the prior art, the sample may be illuminated for longerthan necessary across all regions to avoid data paucity. Because reconstruction involves principally acts of image / mosaic addition, a simplifiedprocessor can be used such as a field programmable gate array (FPGA). This reduceshardware costs and, because of the high processing speed of FPGAs, leads to theaforementioned real-time reconstructed image giving valuable live feedback to the user. Because the reconstruction grid (400) effectively acts as storage for each selected kernel images (e.g.310b, e, h), there is no requirement to store any data associated with theevent driven detector (200). In particular, there is no need to store a record of thedetected event (210a, 210b, 210c). Because an event-driven detector is used, no conventional full frame images are generated, which significantly reduces data handling requirements. Since each detected event (210a, 210b, 210c) is used to select a pre-calculated kernel image (e.g.310b, e, h from 310 a to i) which is added to thereconstruction grid (400), there is no need to store any intermediate reconstructedimages; as mentioned elsewhere, the reconstruction grid (400) effectively acts asstorage. Thus, data storage requirements are reduced or essentially abolished, apartfrom the reconstruction grid (400) and / or reconstructed image.According to one aspect, a method for interactively generating a reconstructed image(500) from electron diffraction data of a test sample (100), comprising:- sending an instruction to an EM to direct a test electron probe to a scan grid vectorlocation (100a, 100b, 100c, ^^⃗), such that the electron probe interacts with the test sample (100) and generates an exit wave (120a);- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c) caused by the generated exit wave (120a);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) from a kernel image library (300) comprising a plurality of kernel images (e.g. 310a to i), wherein: -a kernel image (e.g.310b, e, h) is a pre-calculated mosaic for adding to areconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500);- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400)at a collection of reconstruction grid vector locations (^⃗) centered on the scan grid vectorlocation (^^⃗) that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) on which the kernel image (310b, 310e, 310h) was selected,- repeating (multiple times) the sending, receiving, selecting and adding, wherein for atleast one repetition (preferably for multiple repetitions), a different scan grid vector location (100a, 100b, 100c, ^^⃗) is used or sent, thereby gradually mosaicking and rendering the reconstructed image (500). The comparative steps described herein for the method for generating a reconstructedimage (500) may be applied mutatis mutandis to the method for interactively generatinga reconstructed image (500). Optionally, the reconstruction grid (400) is displayed on a display (such as a computer display, display of a smart device). Optionally, the method may receive an input of the different scan grid vector location (100a, 100b, 100c, ^^⃗). The display and input allow a user or machine to interactively monitor the progress of the reconstruction grid (400) generation.Where one or more locations of the reconstruction grid (400) show deficiency or needimproving, the test electron probe (205) can be moved to the corresponding scan gridvector location (100a, 100b, 100c, ^^⃗), thereby adding additional kernel images ormosaics to the deficient location or location in need of improvement of the reconstructiongrid (400). This means that only areas that require additional data collection areilluminated, thereby reducing beam damage. In the prior art, the whole sample area is beam scanned in the event a region of the reconstruction grid (400) shows a deficiency, which damages all area equally, and causes a distortion in results of subsequent acquisitions.According to one aspect, the location of the reconstruction grid (400) showing deficiencyor needing improving is identified from the display, and the input provided directs the test electron probe (205) to the scan grid vector location (100a, 100b, 100c, ^^⃗) corresponding to the deficient location or location in need of improvement of the reconstruction grid (400). It is understood that the location of the reconstruction grid (400) showing deficiency or needing improving can be identified automatically (e.g. using machine- learning protocols) or based on a user input. The electron diffraction data is acquired using an electron microscope (EM). The EM may be any device emitting an electron beam (test electron probe (105)) capable ofproducing ED data of a test sample. Examples of EMs are a scanning electronmicroscope (SEM), or a scanning transmission electron microscope (STEM). Examplesof EM providers include Zeiss, Tescan, JEOL, Hitachi, ThermoFisher Scientific. Standard methods for acquisition of the ED data of a sample are known in the art, for instance,using standard protocols of the EM, and / or from textbook teachings such as WilliamsD.B. and Carter C.B. “Transmission Electron Microscopy”, Springer New York NY, 2009. The test electron probe (105) is preferably a convergent beam, generating a convergentbeam electron diffraction (CBED) pattern on the count grid (205) of the event-drivendetector. Typically an EM comprises: -an electron source;- a focusing unit configured for forming the electrons into a test electron probe(105); -a deflector unit configured for adjusting the location of the test electron probe(105) on the test sample (100) or a scan grid (120);- one or more event-driven detectors configured for detecting electrons resultingfrom interaction of the test electron probe (105) with the test sample on the scan grid (120). The one or more event-driven detectors (200) is positioned after thetest sample for collecting the ED data;- a controller configured for controlling the EM;- a memory configured for storing computer readable instructions for instructingthe controller to operate the EM according to the present disclosure; -a test sample stage for adjusting the location of the sample or the scan grid(120) with respect to the beam; and- a vacuum chamber and attached vacuum pumps and vacuum sensors.The event-driven detector (200) detects events resulting from the test electron probe(105) having interacted with the test sample (110) on the scan grid (120), resulting in atest event-detected count (TEDC). A TEDC event arises from an interaction between thetest electron probe (105) and the sample (100) at a scan grid vector location (^^⃗) of thescan grid (120). More specifically, the TEDC results from an exit wave of electronsemitted after the test sample (100) striking the count grid (210) at the count grid vectorlocation(s) (^^⃗^). The TEDC is created only when an event is present, in which case it isaccorded a count grid vector location (^^⃗^). The TEDC does not have a gradual amplitude and may be represented as a “one” or “yes” or other binary indication. Asunderstood in the art, the event-driven detector (200) does not output a frame-basedread-out. Regions of the count grid (210) that are not struck with the exit wave do notform a part of an output of the event-driven detector (200). The event-driven detector(200) outputs an event-based read-out. Only regions of the count grid (210) that arestruck with the exit wave form a part of an output of the event-driven detector (200). Thecount grid (210) is located after the test sample (110). Any suitable present or future event-driven detectors may be used. Event-driven detectors are known in the art. Examples of providers include Advacam, Amsterdam Scientific Instruments and Quantum Detectors. The test electron probe (205) is a convergent focused beam that creates a convergent- beam electron diffraction (CBED) pattern on the count grid (205) of the event-driven detector. The test electron probe moves in a scanning pattern over the scan grid (120), and dwells at each scan grid vector location (^^⃗, 100a, 100b, 100c) for a dwell time, and during the dwell time the one or more test event-detected counts (TEDC), is detected by the event driven detector (200). In obtaining the ED data at multiple different scan positions (^^⃗), coils of the deflector unitcontrolling the lateral position of the test electron probe (105) are typically set in ascanning pattern. Types of scanning pattern include Raster scanning, Snake-pattern scanning, Random position scanning, or user-controlled scanning. By raster scan, it is meant that the test electron probe (105) is controlled to move across the sample in anarray of parallel sweeps. By user-controlled scanning, it means that the user is free tomove the scan test electron probe (105) in any direction. User controlled scanning allowsthe user to “paint” onto the reconstruction grid in sweeps in real-time; regions that are poorly resolved may receive repeated sweeps of the test electron probe (105) ascontrolled by the user. The test electron probe (105) is preferably blanked duringrepositioning between two spatially separated (non-adjacent) different scan grid vector locations (100a, 100b, 100c, ^^⃗).The ED data comprises a plurality of test event-detected counts (TEDC), each TEDCrepresented as at least one count grid vector location (^^⃗^) (x,y co-ordinate) on the countgrid (210) of the event-driven detector (210) together with an indication of the scan gridvector location (^^⃗) at which the TEDC occurred. The TEDC does not have a gradualamplitude, and may be represented as a “one” or “yes” or other binary indicator, or justby the presence of the at least one x,y coordinate in the ED data. The time of arrival ofthe event on the event-driven detector (200) does not need to be recorded or present inthe ED data. The TEDC is created only when an event is present, in which case it isaccorded a count grid vector location (^^⃗^).The detection event is typically an activation of one pixel on the count grid (210) at thecount grid vector location (^^⃗^). It is foreseen that the detection event may be a cluster ofadjacent (next-to) pixels on the count grid (210) including the count grid vector location(^^⃗^). The ED data is typically produced at suitable beam energy for generation of the ED data. The beam energy may have an acceleration voltage of 1 to 300 keV, preferably 30 to 300 keV. Since the present disclosures allow ED data to be collected using a lower beam energy, the EM may be capable of providing an EM beam having an energy of 1 keV to30 keV. A meaningful ED dataset may be obtained with a beam dose of 50 to 106electrons received per Å2 in the sample region. This beam dose is typically achieved withdwell times of 1 µs to 1 ms per scan points.A scan grid vector location (100a, 100b, 100c, ^^⃗) may be different and visited only onceby the test electron probe (105). A scan grid vector location (100a, 100b, 100c, ^^⃗) maybe different and visited at least twice by the test electron probe (105). Multiple visitations at a lower dose per visited point result in a reduction in beam damage compared with only one beam visitation per point and higher dose per visited point. Each selected kernel image (310b, 310e, 310h) is added to the reconstruction grid (400)at a collection of reconstruction grid vector locations (^⃗) on the reconstruction grid (400).The vector locations of the collection of reconstruction grid vector locations (^⃗) representa contiguous set of pixels (e.g. a square box) having a corresponding (e.g. same orscaled up or down) dimension to the selected kernel image (310b, 310e, 310h). Byadded or adding, it is meant that pixel intensities from the selected kernel image (310b,310e, 310h) are combined (e.g. by summation) with existing pixel intensities on thereconstruction grid (400) at the locations where the selected kernel image (310b, 310e,310h) is placed on the reconstruction grid (400). Adding each selected kernel image (310b, 310e, 310h) gradually mosaics and renders the reconstructed image (500). Mosaics or mosaicking refers to a process of adding a plurality of selected kernel images (310b, 310e, 310h) (mosaics) at different andsometimes the same reconstruction grid vector locations (^⃗) over time, thereby graduallyrendering the reconstructed image (500). The kernel image (or mosaic) has a size that isa portion (smaller fraction) of the size of the reconstruction grid (400). For instance, thekernel image (or mosaic) has an area that is 50% or less than an area of thereconstruction grid (400). This area ratio is dependent both on the kernel image size andthe number of scan points visited (e.g. the total size of the scan window).The kernel image size is preferably user-defined. Preferably, it represents an area, in the beam-exposed specimen, which is considered as participating in a scattering of the electrons by the test electron probe. It is thus a chosen metric for the real-space extension of the electron probe. It may be expressed in nanometers or pixels. A kernel image size may typically be 9x9 to 41x41 pixels. The adding to the reconstruction grid (400) may be initial, meaning that the selected kernel image (310b, 310e, 310h) or mosaic on the reconstruction grid (400) occupies aposition that is that no existing kernel image has occupied and is not adjacent,overlapping or superimposing with an existing kernel image.An existing kernel image on the reconstruction grid (400) is a selected kernel imageadded previously to the reconstruction grid (400) during the gradual mosaicking andrendering of the reconstructed image (500).The adding to the reconstruction grid (400) may be adjacent, meaning that the selectedkernel image (310b, 310e, 310h) or mosaic is positioned next to an existing kernel imageon the reconstruction grid (400) (no overlap). The adding to the reconstruction grid (400) may additionally or alternatively beoverlapping, meaning that the selected kernel image (310b, 310e, 310h) or mosaic ispositioned so that one or more of its edges overlaps with a part of one or more one or more existing kernel images on the reconstruction grid (400). The adding to the reconstruction grid (400) may additionally or alternatively besuperimposing, meaning that the selected kernel image (310b, 310e, 310h) or mosaic ispositioned exactly over one or more existing kernel images on the reconstruction grid(400). In FIG. 1, for instance, three kernel images (310b, 310e, 310h) are added bysuperimposing addition to the reconstruction grid (400). After the selected kernel image (310b, 310e, 310h) or mosaic is added to the reconstruction grid (400), the reconstruction grid becomes updated. Only a portion of the reconstruction grid (400) where the kernel image (310b, 310e, 310h) or mosaic has been added may become updated. Multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500).The reconstruction grid vector locations (^⃗) at which the selected kernel image (310b,310e, 310h) is added to the reconstruction grid (400) are determined from the scan grid vector location (^^⃗) that is a location where a test electron probe (105) interacts with thetest sample (100) on a scan grid (120). Typically, the reconstruction grid vector location(^⃗) is determined from a proportional relation with the scan grid vector location (^^⃗),optionally allowing for a border (e.g. blank or black) to be present in the reconstructiongrid (400). The proportional relationship may be expressed according to Eq. 1:(^⃗) = ^^^^^.(^^⃗) + ^^^^^^^,^[Eq.1] where: (^⃗) is the reconstruction grid vector location to be determined; (^^⃗) scan grid vector location;^^^^^ is a scaling factor constant (scaling may be up or down). It is one in absence ofscaling.^^^^^^^,^ is an offset constant, optionally allowing presence of a border. It is zero inabsence of a border. By adding pixels and mosaicking the plurality of selected kernel images (310b, 310e,310h), the signal to noise of the reconstruction grid (400) the reconstructed image (500)is increased.The kernel image library (300) comprises the plurality of pre-calculated kernel images(310a to i). Each kernel image (310a to i) of the plurality is a mosaic for adding to the reconstruction grid (400) of the reconstructed image (500). Each kernel image (310a to i)is simulated from a single simulated event-detected count having a simulated count gridvector location (^^⃗^) on a simulated count grid of a simulated count detector.Each kernel image (310a to i) in the kernel image library (300) is linked with the simulated count grid vector location (^^⃗^) (320a to 320i) that generated the kernel image(310a to i) or mosaic. Thus, by querying the kernel image library (300) with a test countgrid vector location (^^⃗^), the kernel image (310a to i) linked to the equivalent simulatedcount grid vector location (^^⃗^) can be selected. Thus, kernel image (310b, 310e, 310h)is selected when (^^⃗^) matches (^^⃗^). The kernel image (310a to i) may represent a partial measurement of an electrostaticpotential (μ) map weighted by an interaction constant (σ), of a charge density (ρ) mapweighted by the same interaction constant (σ), or of the real or imaginary parts of a transmission function (T) map. This transmission function is the complex exponential of the electrostatic potential (μ) weighted by the interaction constant (σ).By pre-calculated, it is meant that the kernel image library (300) is generated prior to theacquisition of the electron diffraction data or prior to the data collection session. The priorto may be, for instance, within a few (e.g.1 to 3) days, hours or minutes before theacquisition of the electron diffraction data. The kernel image library (300) does notdepend on the sample and, hence, can be retrieved from a database of stored kernelimage libraries (300). The prior to may be, for instance, longer than a few (e.g.1 to 3)days, hours or minutes before the acquisition of the electron diffraction data.Each kernel image (310a to i) in the kernel image library (300) is simulated from a single simulated count having a simulated count grid vector location (^^⃗^). The simulation comprises using analytical ptychography, whose definition is here extended to includethe iCoM method. The simulation does not require the scan grid vector locations (100a,100b, 100c, ^^⃗). The analytical ptychography is preferably one of integrated centre of mass (iCoM), side- band integration (SBI), alternatively known as single side-band (SSB) rconstruction, or Wigner distribution deconvolution (WDD). Those methods are known in the art. iCOM has been first described “Phase contrast STEM for thin samples: Integrateddifferential phase contrast”, Lazić et al, Ultramicroscopy, Volume 160, January 2016,pages 265-280 It is important to notethat this method is distinct from integrated differential phase contrast (iDPC), which was first described in this publication as well.SBI has been first described in “Experimental tests on double-resolution coherentimaging via STEM”, Rodenburg, et al, Ultramicroscopy, Volume 48, Issue 3, March 1993, Pages 304-314 (https: / / doi.org / 10.1016 / 0304-3991(93)90105-7).WDD has been first described in “The theory of super-resolution electron microscopy viaWigner-distribution deconvolution”, Rodenburg and Bates, Philosophical Transactions of the Royal Society of London. Series A: Physical and Engineering Sciences, 15 June 1992, Volume 339, Issue 1655 (https: / / doi.org / 10.1098 / rsta.1992.0050).Each kernel image (310a to i) may represent a partial measurement of an electrostaticpotential (μ) map weighted by an interaction constant (σ), of a charge density (ρ) map weighted by the same interaction constant (σ), or of the real or imaginary parts of atransmission function (T) map, as defined elsewhere in this document. The reconstructedimage (500) may accordingly be an electrostatic potential (μ) map weighted by an interaction constant (σ), a charge density (ρ) map weighted by the same interaction constant (σ), or the real or imaginary parts of a transmission function (T) map of the sample. To generate a kernel image (310a to i) in the kernel image library (300), the usualformalisms for the iCoM, SBI and WDD methods are expanded by describing thesimulated diffraction data as a sum of Dirac functions, one for each simulated count. Thepresence of those Dirac functions permits the mathematical simplification of integrals and Fourier transforms, which otherwise would have to be explicitly performed on a complete four-dimensional electron diffraction dataset. Each analytical ptychography process is consequently reduced to a summation of kernel images, each being pre-calculated and reduced in real-space by a given kernel image size, hence permitting fast treatment and economy of memory bandwidth.The simulation of each kernel image preferably uses one or more (preferably all) of:- one or more recording condition parameters (acceleration voltage ^ (in kV), semi-convergence angle ^ (expressed in mrad), scan step size (in nm);- a calibration of the (^^⃗^) coordinates of the count grid (205) on the event-driven detector(200).- a desired kernel size (in nm);- a Wiener number ε (e.g. small number used to avoid divisions by zero, typically takenbetween 10-4and 10-8, and acting as a noise filtering parameter); Other optional parameters may be chosen for band-pass filtering, aberration correction, incoherence compensation, contrast transfer function (CTF) compensation and modulation transfer function (MTF) compensation. The image kernel library (300) does not depend on the sample.The kernel image library (300) is stored in memory accessible to a standard computersystem configured to generate the reconstructed image (500). The standard computer system is configured to carry out one or more of the methods described herein. Thestandard computer system may be, for instance, an Intel Architecture IA-32 basedcomputer system 2. The method(s) may be implemented as programming instructions ofone or more software modules stored on non-volatile (e.g. hard disk or solid-state drive) storage associated with the corresponding computer system. However, as mentionedelsewhere herein, at least some of the steps, especially the selecting a kernel imageand / or generating of the reconstructed image, could alternatively be implemented, either in part or in its entirety, as one or more dedicated hardware components, such as gate configuration data for one or more field programmable gate arrays (FPGAs), or as application-specific integrated circuits (ASICs), for example.After an event has been detected by the event driven detector (200), the test count gridvector location (^^⃗^) of the event is queried against the kernel image library (300). Thekernel image (310a to i) of the kernel image library (300) having an equivalent simulated count grid vector location (^^⃗^) is selected. The selected kernel image (310a to i) is added to the reconstruction grid (400). The method or system described herein may be used for real-time generation of the reconstructed image (500) from the electron diffraction data. By real-time, it is meant that a detected event (210a, 210b, 210c) on the count grid (210) of the event driven detector(200) triggers the selecting of a kernel image (e.g.310b, e, h) from the kernel imagelibrary (300). The selecting of a kernel image (e.g.310b, e, h) occurs before the testelectron probe (105) has completely scanned the scan grid (120). Preferably, theselecting of the kernel image (e.g.310b, e, h) occurs before the test electron probe (105)moves to the next vector position (^^⃗) of the scan grid (120). Real-time may mean that theED data generated during the data collection session is not stored in storage memorysuch as a hard drive. It is foreseen that multiple detected events (210a, 210b, 210c) maybe buffered e.g. held in transitory storage of limited capacity during real-time processing.Real-time generation allows one or more locations of the reconstruction grid (400)showing deficiency or need improving to be identified during acquisition of the ED data,allowing the test electron probe (205) to be moved to the corresponding scan grid vectorlocation (100a, 100b, 100c, ^^⃗), for adding of additional kernel images or mosaics to thedeficient location or location in need. As mentioned elsewhere herein, each kernel image (310a to i) is simulated from a singlesimulated event-detected count having a simulated count grid vector location (^^⃗^) on asimulated count grid of a simulated count detector. What follows is a detailed descriptionof the generation of a pre-calculated kernel image (310a to i) for the kernel image library(300).A simulated diffraction dataset contains a collection of simulated event-detected countsdetected at simulated scan positions r⃗ ^ ^ , by pixels at spatial frequencies q^⃗^^⃗^ ^ on thesimulated count detector, e.g. a collection of indices d for each indice c.A(q^⃗ ) represents the aperture inserted in the focal plane of the microscope. It is equal to 1for spatial frequencies q^⃗ fulfilling ‖q^⃗ ‖ ≤ q^, and 0 otherwise. q^ is a spatial frequencylimit given by q^= ^, with α the semi-convergence angle of the illumination, and λ theelectron wavelength, which itself depends on the acceleration voltage U.A^^(r⃗) is a kernel limitation term, including a Hann window. The limitation radius, orkernel radius, is chosen by the user and expressed in nm. This distance directlytranslates into a number of pixels for the kernel image size.A^^(q^⃗ ) is a band-pass filtering window. Frequency limits are chosen by the user. Thisband-pass filtering window can alternatively be replaced by a frequency weightingfunction, e.g. by assigning to it a value taken between 0 and 1 and depending directly onthe spatial frequency q^⃗ .ε is a Wiener parameter, chosen by the user.p is a CTF compensation parameter, taken between 0 and 1 and chosen by the user. Δr^is the effective source size to be compensated (spatial incoherence correction), Δf^is the focus spread to be compensated (temporal incoherence correction). Δr^and Δf^are provided by the user as single quantities (expressed in nm)χ^(q^⃗ ) is the aberration function to be corrected. It is given by with the parameters C^,^(expressed in nm) and β^,^(expressed in radians) beingprovided by the user for specific couples [a, b] of integers, provided by the user as well.∠q^⃗ is the angular orientation of a spatial frequency q^⃗ and ‖q^⃗ ‖ its modulus.M(r⃗) represents the modulation transfer function of the camera. It is provided by the userin case its correction is desired. Otherwise, it is left equal to 1.F and F^^ respectively represent a Fourier transform and an inverse Fourier transform.The Fourier transform of an arbitrary function f(r⃗) is given by And the inverse Fourier transform of an arbitrary function g(q^⃗ ) is given by CTF compensation term: Side-Band Integration (SBI)Electrostatic potential:Charge Density: According to one aspect: B^^^Q^⃗ ; q^⃗ ^ and B^^^Q^⃗ ; q^⃗ ^ are double-overlap functions, given byB^^^Q^⃗ ; q^⃗ ^ = A(q^⃗ ) A^q^⃗ − ^ Q^⃗ ^ ^1 − A^q^⃗ + ^ Q^⃗ ^^B^^^Q^⃗ ; q^⃗ ^ = A(q^⃗ ) A^q^⃗ + ^ Q^⃗ ^ ^1 − A^q^⃗ − ^ Q^⃗ ^^φ^^^Q^⃗ ; q^⃗ ^ and φ^^^Q^⃗ ; q^⃗ ^ illumination correction terms, may be given by^^^^(^⃗)^(^⃗)^^^^^⃗^^^^⃗ ^^ More preferably, φ^^^Q^⃗ ; q^⃗ ^ and φ^^^Q^⃗ ; q^⃗ ^ illumination correction terms, are given by^^^^(^⃗)^(^⃗)^^^^^⃗^^^^⃗ ^^ CTF compensation term: Wigner Distribution Deconvolution (WDD)Imaginary part:^q^⃗^^; r⃗ − r⃗^ ^ ^ ^⃗^Real part: With Illumination convolution term: Γ^^Q^⃗ ; ^ R⃗^ = F^^^ A(q^⃗ ^) A^q^⃗ ^ + ^ Q^⃗ ^ θ^q^⃗ ^ ; ^ Q^⃗ ^ ^^^R⃗^Illumination correction term may be: correctionterm more preferably is:θ q^⃗ ^ Following their initial calculation through the formulas given above, the libraries may beemployed to generate images through the processes described elsewhere in thisdocument. The spatial frequency grid (q^⃗ ) corresponds to the simulated count grid vectorlocations (^^⃗^). They themselves coincide with the spatial frequency coordinates of thecount grid of the event-driven detector, having been calibrated before the calculation.The real-space grid (r⃗) is the reconstruction grid. The notation r⃗ − r⃗^ ^ used previously indicates the centering of the kernel image, having first been selected by correspondence of (^^⃗^) with (^^⃗^), on a scan position r⃗ ^ ^ .The ED data is collected during a data collection session. In a data collection session,the sample remains in the same position relative to a sample holder. Provided herein is an EM (and computing device or system), configured to carry out themethod described herein. The EM may be capable of providing an EM beam having anenergy of 1 keV to 300 keV, preferably 30 to 300 keV. Since the present disclosures allow ED data to be collected using a lower beam energy, the EM may be capable ofproviding an EM beam having an energy of 1 keV to 30 keV.The presently described method is a computer-implemented method. The method maybe performed by a standard (generic) computer system as described elsewhere herein.Further provided is a computing device or system configured for performing a method as described herein, or a part thereof. Further provided is a computer program or computer program product having instructions which when executed by a computing device or system cause the computing device or system to perform a method as described herein, or a part thereof. Further provided is a computer readable medium having stored thereon a computer program (product) having instructions which when executed by a computing device or system cause the computing device or system to perform (each of the steps of) the method as described herein, or a part thereof. Further provided is a data stream which is representative of a computer program or computer program product having instructions which when executed by a computing device or system cause the computing device or system to perform (each of the steps of) the method as described herein, or a part thereof. It is understood that the presently-described method may alternatively be performed by an electron microscope incorporating a standard computer system.Further provided is an electron microscope configured for performing a method asdescribed herein, or a part thereof. Further provided is a computer program or computer program product having instructionswhich when executed by the electron microscope cause the electron microscope toperform a method as described herein, or a part thereof. Further provided is a computer readable medium having stored thereon a computer program (product) having instructions which when executed by a electron microscopecause the electron microscope to perform (each of the steps of) the method as describedherein, or a part thereof. Further provided is a data stream which is representative of a computer program orcomputer program product having instructions which when executed by the electronmicroscope cause the electron microscope to perform (each of the steps of) the methodas described herein, or a part thereof.

Claims

Claims1. A method for generating a reconstructed image (500) from electron diffraction data ofa test sample (100), comprising:- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) - a selected kernel image (310b, 310e, 310h) - from a kernel image library(300) comprising a plurality of kernel images (e.g.310a to i), wherein:- each kernel image (e.g.310a to i) of the kernel image library (300) is pre-calculated and is a mosaic for adding to a reconstruction grid (400) of the reconstructed image (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500); and- generating the reconstructed image (500), comprising:- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400)thereby gradually mosaicking and rendering the reconstructed image (500).

2. The method according to claim 1, wherein each kernel image (310a to i) of the kernelimage library (300):- has a size that is a portion of the size of the reconstruction grid (400);- is simulated from a single simulated event-detected count having a simulated countgrid vector location ^^⃗^on a simulated count grid of a simulated count detector.

3. The method according to claim 1 or 2, wherein:- the electron diffraction data further comprises a plurality of different scan grid vectorlocations (100a, 100b, 100c), each scan grid vector location being a location of a test electron probe that interacts with the test sample (100) at vector position ^^⃗ of the scan grid (120) thereby generating an exit wave (120a) from the test sample (100);- each test event-detected count (210a, 210b, 210c) of the plurality of test event-detected counts (210a, 210b, 210c) has a test count grid vector location ^^⃗^ on a countgrid (205) of the event-driven detector (200);- the test count grid vector location ^^⃗^ is linked in the electron diffraction data to thescan grid vector location (100a, 100b, 100c,) that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) at the test count grid vector location ^^⃗^;- the selected kernel image (310b, 310e, 310h) is selected when ^^⃗^ matches ^^⃗^;- the adding of each selected kernel image (310b, 310e, 310h) to the reconstruction grid(400) is at a collection of reconstruction grid vector locations, ^⃗, centered on the scangrid vector location that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) on which the selected kernel image (310b, 310e, 310h) was selected.

4. The method according to claim 3, wherein the adding of each selected kernel image(310b, 310e, 310h) to the reconstruction grid (400) comprises a step of summingindividual pixel intensities of the added selected kernel image (310b, 310e, 310h) ateach reconstruction grid vector location ^⃗ with previous individual pixel intensities at thesame reconstruction grid vector locations ^⃗.

5. The method according to claim 3 or 4, wherein the adding of each selected kernelimage (310b, 310e, 310h) to the reconstruction grid (400) at the reconstruction gridvector location ^⃗ is adjacent, overlapping, or superimposing, wherein:- the adjacent adding comprises positioning the selected kernel image (310b,310e, 310h) next to an existing kernel image on the reconstruction grid (400 -the overlapping adding comprises positioning the selected kernel image (310b,310e, 310h) so that one or more of its edges overlaps with a part of one or more existing kernel images on the reconstruction grid (400); -the superimposing adding comprises positioning the selected kernel image(310b, 310e, 310h) exactly over one or more existing kernel images on the reconstruction grid (400); and -the existing kernel image on the reconstruction grid (400) is a selected kernelimage added previously to the reconstruction grid (400) during the gradualmosaicking and rendering of the reconstructed image (500).

6. The method according to any one of claims 1 to 5, wherein after the selected kernel image (310b, 310e, 310h) is added to the reconstruction grid (400), the reconstruction grid becomes updated, and only a portion of the reconstruction grid (400) where theselected kernel image (310b, 310e, 310h) has been added becomes updated.

7. The method according to any one of claims 1 to 6, wherein the size of each kernelimage (310a to i) of the kernel image library (300) is equal to or less than 50% of thesize of the reconstruction grid (400) by area.

8. The method according to any one of claims 2 to 7, wherein the simulation of eachkernel image (310a to i) in the image library (300) from a single simulated count havinga simulated count grid vector location ^^⃗^comprises analytical ptychography.

9. The method according to claim 8, wherein: -the analytical ptychography is integrated centre of mass (iCOM), or side-bandintegration (SBI), or Wigner distribution deconvolution (WDD); and -the simulation of the kernel image (310a to i) in the image library (300)represents a partial measurement of an electrostatic potential (μ) map weighted by an interaction constant (σ), of a charge density (ρ) map weighted by the same interaction constant (σ), or of the real or imaginary parts of a transmission function (T) map, wherein the transmission function is the complex exponential of the electrostatic potential (μ) weighted by the interaction constant (σ).

10. The method according to any one of claims 1 to 9 wherein the generating of thereconstructed image (500) is performed in real-time.

11. A method for interactively generating a reconstructed image (500) from electrondiffraction data of a test sample (100), comprising the method of any one of claims 1 to10, wherein: - the method further comprises sending an instruction to an electron microscope (EM)to direct a test electron probe to a scan grid vector location (100a, 100b, 100c), suchthat the electron probe interacts with the test sample (100) and generates an exit wave(120a);- in the step of:receiving the electron diffraction data of the test sample (100) comprising a plurality of test event-detected counts (210a, 210b, 210c) caused by the generated exit wave (120a),said electron diffraction data of the test sample (100) is generated by the electronmicroscope (EM);- in the step of:adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400) thereby gradually mosaicking and rendering the reconstructed image (500), each selected kernel image (310b, 310e, 310h) is added to the reconstruction grid (400)at a collection of reconstruction grid vector locations (^⃗) centered on the scan grid vectorlocation that generated the exit wave (120a) causing the test event-detected count(210a, 210b, 210c) on which the selected kernel image (310b, 310e, 310h) was selected;- the method further comprises:repeating the sending, receiving, selecting and adding, wherein for at least onerepetition, a different scan grid vector location (100a, 100b, 100c) is used, therebygradually mosaicking and rendering the reconstructed image (500).

12. A method for interactively generating a reconstructed image (500) from electrondiffraction data of a test sample (100), comprising:- sending an instruction to an electron microscope (EM) to direct a test electron probe toa scan grid vector location (100a, 100b, 100c,), such that the electron probe interactswith the test sample (100) and generates an exit wave (120a);- receiving the electron diffraction data of the test sample (100) comprising a plurality oftest event-detected counts (210a, 210b, 210c) caused by the generated exit wave (120a);- selecting, for each test event-detected count (210a, 210b, 210c), a kernel image (e.g.310b, e, h) - a selected kernel image (310b, 310e, 310h) - from a kernel image library(300) comprising a plurality of kernel images (e.g.310a to i), wherein:- each kernel image (e.g.310a to i) of the kernel image library (300) is a pre-calculated mosaic for adding to a reconstruction grid (400) of the reconstructedimage (500), wherein multiple selected added mosaics are capable of mosaicking and rendering the reconstructed image (500);- adding each selected kernel image (310b, 310e, 310h) to the reconstruction grid (400)at a collection of reconstruction grid vector locations, ^⃗, centered on the scan grid vectorlocation that generated the exit wave (120a) causing the test event-detected count (210a, 210b, 210c) on which the kernel image (310b, 310e, 310h) was selected,- repeating the sending, receiving, selecting and adding, wherein for at least onerepetition, a different scan grid vector location (100a, 100b, 100c) is used, therebygradually mosaicking and rendering the reconstructed image (500).

13. The method according to claim 12, incorporating the subject matter of any one ofclaims 2 to 10.

14. A method according to any one of claims 1 to 13 that is a computer-implemented method.

15. A computing device or system configured for performing the method accordingtoclaim 14.

16. A computer program or computer program product having instructions which when executed by a computing device or system cause the computing device or system to perform the method according to claim 14.

17. A computer readable medium having stored thereon a computer program having instructions which when executed by a computing device or system cause thecomputing device or system to perform the method according to claim 14.

Images