Method for acquiring images from a pseudoperiodic coding target for micro-localization, and associated device

Abstract

The present invention relates to a method for acquiring images by a matrix image detector of a pseudoperiodic pattern used for micro-localization, comprising an elementary periodic grid. The matrix image detector, controlled by a controller connected to a processing unit, has acquisition parameters, including at least one parameter for reducing pixels into rows and / or columns.According to the invention, the method comprises the following steps: - determination (E1) of an elementary spatial period in pixels of the elementary periodic grid, - calculation (E2) of at least one pixel reduction parameter in lines and / or in line, as a function of the elementary spatial period, - adjustment (E3) of at least one acquisition parameter of the matrix image detector by the controller according to the calculation to obtain reduced acquisition parameters, - acquisition (E4) of a reduced image of the pseudoperiodic pattern by applying the reduced acquisition parameters to the matrix image detector. Figure for abstract: Figure 5.

Description

Title of the invention: Method for acquiring images of a pseudoperiodic coding target for micro-localization, and associated device Technical field of the invention

[0001] The present invention generally relates to a spatial location method for an imaging system.

[0002] It relates more particularly to a method for acquiring images of a pseudoperiodic coding target for micro-localization, this allowing position and orientation identification for a microscope or observation instrument associated with a system for moving the observed object, in order to determine the performance of precision and repeatability of positioning at a sub-micrometric, or even nanometric, level, optimized in terms of acquisition and processing time.

[0003] It further relates to an associated device. State of the art

[0004] The question of positioning accuracy and its stability over time for a microscopy system has become an important concern for the observation and / or study of samples at a micrometric scale. Indeed, thanks to the advent of increasingly efficient and automated microscopy systems, observations can be carried out for hours, or even entire days, at spatial resolutions of the order of a hundred nanometers.

[0005] In order to facilitate these observations, which can be multi-scale and multimodal, it has been proposed to use micrometric or even nanometric precision positioning systems. These so-called micro-localization systems make it possible to obtain spatially correlated measurements on a given sample, which is observed using several measuring instruments, and the location of a measurement point on this sample, with a precision lower than the pixel of the imaging system of the measuring instrument.

[0006] Document EP3458896 describes such a positioning system. It proposes a micro-localization device for an imaging instrument, the device defining a spatial coordinate system specific to the sample to which it is attached, and the associated positioning method. This positioning method is based on the acquisition by an imaging instrument, followed by the analysis of an image of the micro-localization device, which takes the form of a pseudoperiodic encoding target. Algorithmic processing of the acquired image finally makes it possible to deduce positioning and / or orientation coordinates.

[0007] The publication by July-Andrea Galeano-Zea, Patrick Sandoz, Emilie Gaiffe, Jean-Luc Prétet & Christiane Mougin (2010) Pseudo-Periodic Encryption of Extended 2-D Surfaces for High Accurate Recovery of any Random Zone by Vision, International Journal of Optomechatronics, 4:1, 65-82, DOI: 10.1080 / 15599611003660395 also describes a micro-localization system, in the form of a target comprising a pseudoperiodic pattern, and of which an acquired image is used in the context of algorithmic processing, in order to deduce positioning and / or orientation coordinates.

[0008] Such micro-localization devices have been used in particular to observe the phenomena of parasitic vibrations and / or drift of the region of interest which can affect the image quality. Indeed, vibrations and / or drifts, even micrometric due for example to mechanical instability of the measuring instrument, generate blurred images and / or images not centered on the phenomenon of interest.

[0009] However, these observations require a high imaging rate, in order to observe transient phenomena or vibrations.

[0010] Image acquisition and processing when carried out on a microlocation device in order to retrieve positioning and / or orientation information, as described in the prior art, are therefore faced with two significant drawbacks when high observation rates are required. The first drawback relates to the limitation of the imaging frequency, the second is linked to the processing time and / or the size of the data required for the algorithmic processing for an evaluation of the resulting position. Presentation of the invention

[0011] In order to remedy the aforementioned drawbacks of the state of the art, the present The invention proposes a method for acquiring images of a pseudoperiodic coding target for micro-localization.

[0012] More particularly, according to the invention, there is proposed a method for acquiring images using a matrix image detector, comprising N1 rows and M1 columns of pixels, the matrix image detector being included in an imaging instrument, the matrix image detector being controlled by a controller connected to a processing unit, and having acquisition parameters, including at least one parameter for reducing pixels into rows and / or columns; the pseudoperiodic coding pattern comprising an elementary periodic grid in which at least one pattern is arranged. It is thus provided that the method comprises the following steps: - determination by the processing unit of an elementary spatial period in pixels of the elementary periodic grid from a first image of the target acquired by the matrix image detector, - calculation by the processing unit of at least one pixel reduction parameter in rows and / or columns as a function of the elementary spatial period in pixels and a minimum spatial period in pixels, - adjustment of at least one acquisition parameter of the matrix image detector by the controller using at least one pixel reduction parameter in rows and / or columns transmitted by the processing unit, to obtain reduced acquisition parameters, - acquisition of a reduced image of the pseudoperiodic encoding target by applying to the matrix image detector the reduced acquisition parameters resulting from the adjustment of the previous step.

[0013] Thus, thanks to the invention, the acquisition parameters are chosen so as to reduce the number of pixels of the matrix image detector which will be read individually during acquisition, thus reducing a size in bytes (or bytes in English) of the image to be processed by a processing algorithm during the determination of the location. The reduced number of pixels read individually makes it possible to accelerate the imaging rate since the time necessary for the acquisition of an image is thus reduced, while the reduction in the size of the image to be processed reduces the processing time. The reduced acquisition parameters, however, retain all the information necessary for the precise determination of the orientation and / or the position.

[0014] Advantageously, the invention also allows successive acquisitions carried out over long periods of time, thanks to the reduction in the size in bytes of an image, which facilitates the management of available memory space.

[0015] Other non-limiting and advantageous characteristics of the method according to the invention, taken individually or in all technically possible combinations, are the following: - the at least one parameter for reducing pixels into rows and / or columns comprises a parameter for grouping pixels into rows and / or columns, - the at least one parameter for reducing pixels into rows and / or columns comprises a parameter for subsampling pixels into rows and / or columns, - the matrix image detector comprises a recording of at least one pre-established list of predetermined values ​​of the parameter for reducing pixels into rows and / or columns, the method comprising during the step of adjusting at least one acquisition parameter, a sub-step of selecting at least one value in the recording as a function of the result of the calculation step, - during the step of determining an elementary spatial period of the elementary periodic grid, the first image corresponds to an initial image, which is read individually on all of the NI lines and Ml columns of pixels of the matrix image detector, taken one by one, - the following steps are added between the step of determining an elementary spatial period elementary periodic grid and the acquisition step: - determination by the processing unit of at least one dimension in pixels of an elementary periodic grid resulting from the pseudoperiodic coding pattern, - calculation by the processing unit of at least one cropping parameter in rows and / or columns as a function of at least one dimension in pixels of an elementary periodic grid resulting from the pseudoperiodic coding pattern, - adjustment of at least one acquisition parameter of the matrix image detector by the controller with the at least one image cropping parameter in rows and / or columns transmitted by the processing unit, to obtain reduced acquisition parameters. - during the step of determining at least one dimension of an elementary periodic grid from the pseudoperiodic coding pattern, the dimension in pixels of the elementary grid is determined from the value of an integer multiple stored in a storage unit and the elementary spatial period in pixels of the elementary periodic grid, - during the step of determining at least one dimension of an elementary periodic grid from the pseudoperiodic coding pattern, the dimension in pixels of the elementary grid is determined from an image acquired on the matrix image detector, - the step of calculating at least one row and / or column cropping parameter takes into account an orientation of the elementary periodic grid relative to two axes of the matrix image detector, - the at least one value calculated in the step of calculating at least one row and / or column cropping parameter, takes into account the row and / or column pixel reduction parameters calculated in the corresponding calculation step, - the method is adapted so that the step of acquiring an image of the pseudoperiodic coding test pattern by applying reduced acquisition parameters to the matrix image detector requires an acquisition time less than an acquisition time necessary for an image acquisition, the image being read on the entirety of the NI rows and Ml columns of pixels of the matrix image detector, taken one by one.

[0016] The invention also relates to a device for acquiring images of a pseudoperiodic coding target for micro-localization comprising a processing unit and a matrix image detector comprising N1 lines and M1 columns of pixels, and having acquisition parameters including parameters for reducing pixels into lines and / or columns, the matrix image detector being controlled by a controller connected to the processing unit, where it is provided that: - the processing unit is adapted to determine an elementary spatial period in pixels of an elementary periodic grid contained in the pseudoperiodic coding pattern from a first image of the pattern acquired by the matrix image detector, and to calculate at least one pixel reduction parameter in rows and / or columns as a function of the elementary spatial period and a minimum spatial period in pixels, - the controller is adapted to receive at least one pixel reduction parameter in rows and / or columns from the processing unit, - the controller is adapted to adjust at least one acquisition parameter of the matrix image detector as a function of the at least one pixel reduction parameter in rows and / or columns, to obtain reduced acquisition parameters, - the matrix image detector is adapted to acquire a reduced image of the pseudoperiodic encoding pattern by applying the reduced acquisition parameters.

[0017] Of course, the various features, variants and embodiments of the invention may be combined with each other in various combinations to the extent that they are not incompatible or mutually exclusive. Detailed description of the invention

[0018] Furthermore, various other characteristics of the invention emerge from the appended description given with reference to the drawings which illustrate a non-limiting form of embodiment of the invention and where:

[0019] [Fig-1] is a schematic side view of an imaging instrument in position of measurement against a pseudoperiodic encoding target for micro-localization;

[0020] [Fig.2] is a schematic top view of the imaging instrument of the [Fig.l], where the maximum field of view of a matrix image detector is represented in particular;

[0021] [Fig.3] is a representation of an initial image of the pseudo-coded target periodic for the micro-localization of [Fig.l], acquired by the matrix image detector, and corresponding to the field of view of [Fig.2];

[0022] [Fig.4] is a representation of a reduced and cropped image of the coding target pseudoperiodic for the micro-localization of the [Fig.l] acquired by the matrix image detector, after application of the image acquisition method which is the subject of the present disclosure;

[0023] [Fig.5] is a flowchart of the image acquisition method applied to the detector of matrix images for the acquisition of the image of [Fig.4], comprising at least one calculation step;

[0024] [Fig.6] is a flowchart detailing the calculation step of the acquisition process of images presented in [Fig.5];

[0025] In the remainder of the description, the term micro-location refers to a location in position and / or orientation with submillimeter or even micrometer precision. Similarly, by precision in position and / or orientation, we mean both absolute precision (accuracy, in English) and / or repeatability (repeatability in English).

[0026] Furthermore, the movements considered here range from the order of ten centimeters to a few nanometers.

[0027] In [Fig.l], an imaging instrument 1 is schematically represented via a side view. This imaging instrument 1 comprises an imaging optical system 10, a displacement stage 14, a processing unit 18, as well as a matrix image detector 16, which is controlled by a controller 160.

[0028] This imaging instrument 1 corresponds to a microscope, for example an optical microscope, a Raman microscope, an electron microscope, or any other suitable measuring and / or analysis instrument.

[0029] Here, the case of an optical microscope is considered, comprising at least one optical imaging system 10, for example, a microscope objective. This microscope objective has a magnification, ranging for example from 5x to 100x, according to the notation known to those skilled in the art.

[0030] The movement plate 14 makes it possible to place an observation object on a position of interest and / or to return to it and / or to scan a region of interest on this object.

[0031] In order to determine the displacement precision performance as well as the phenomena of parasitic vibrations and / or drift of a region of interest which may affect the displacement stage 14, during an observation made using the imaging instrument 1, a micro-localization device is used here.

[0032] This micro-localization device is for example directly fixed on the displacement stage 14, the precision and reproducibility of which are to be evaluated. Alternatively, it is also possible to fix the micro-localization device 3 on a sample holder, itself placed on the displacement stage 14.

[0033] By "fixed", it is understood that the micro-location device is placed in a fixed position on the displacement plate 14. For this, the micro-location device is possibly glued in a non-permanent manner, for example using an adhesive, or a wax that is solid at room temperature and liquid when hot. Here, it is assumed that the micro-location device is simply placed on the surface of the displacement plate 14, in a position assumed to be fixed relative to the displacement plate 14, even during the movements and / or vibrations thereof.

[0034] The micro-location device has, for example, macroscopic dimensions; here, it is a target of 25 millimeters by 25 millimeters, or even 200 millimeters by 200 millimeters. In particular, the dimensions of the micro-location device 3 are chosen according to the magnitude of the movements that are to be captured. characterize accuracy and reproducibility.

[0035] More specifically, the micro-location device comprises a pseudoperiodic coding pattern 3. Such a pseudoperiodic coding pattern 3 is shown in more detail in [Fig.2], via a front view. This pseudoperiodic coding pattern 3 comprises at least one elementary periodic grid 30.

[0036] Here, a plurality of elementary periodic grids 30 form a regular two-dimensional tiling. In the example of [Fig.2], the different elementary periodic grids 30 have the same size. In one embodiment, this pseudoperiodic coding target 3 comprises a plurality of zones, each of the zones being paved with elementary periodic grids 30 of a size adapted to be viewed using a certain magnification of the imaging instrument 1 and / or a certain measurement and / or analysis method depending on the imaging instrument 1.

[0037] Each elementary periodic grid 30 of the pseudoperiodic coding pattern 3 makes it possible to advantageously code positioning and / or orientation information, as well as information relating to the pseudoperiodic coding pattern 3 and / or to a possible observation object. Document EP3458896 provides a more detailed description of such a pattern, the design of which is not claimed in the present disclosure.

[0038] Here, each elementary periodic grid 30 of the pseudoperiodic coding pattern 3 is delimited by a rectangular border. A space delimited by this border is then divided regularly, that is to say spatially periodic, into boxes. A dimension of these boxes defines an elementary spatial period Tpix, which therefore corresponds to the spatial period of the elementary periodic grid.

[0039] It is considered here that the boxes are square in shape, the elementary spatial period Tpix is ​​therefore valid for both directions of the plane. An elementary spatial period Tpix of distinct value and defined for each of the directions of the plane is also possible.

[0040] Patterns 301 are then arranged in boxes of this elementary periodic grid 30, in order to code the target.

[0041] The encoding is therefore carried out by placing patterns 301 in each box of the elementary periodic grid 30.

[0042] A wide variety of patterns 301 is possible. In the example illustrated in [Fig.2] and detailed in [Fig.3], some of these patterns 301 correspond to a block of 2x2 squares alternately black and white. This block thus corresponds to a checkerboard pattern. Other patterns 301 are also arranged within the elementary periodic grids 30 of [Fig.2] and 3, such as squares joined on sides Tpix by Tpix, that is to say a square occupying the entirety of a box of the elementary periodic grid 30. This square is here black, or white.

[0043] Another alternative consists of encoding the target using an alternation of a pattern 301 taking the form of a white square of side Tpix by Tpix in the center of which is placed a black point and a pattern 301 taking the form of a white square of Tpix by Tpix united.

[0044] The encoding of the pseudoperiodic coding pattern 3 therefore resides in the alternation of patterns 301. Similarly, a method of encoding the pseudoperiodic coding pattern 3 including a multitude of different patterns 301, as in this embodiment shown, also allows the encoding of the pseudoperiodic coding pattern 3.

[0045] Regardless of the pattern 301 chosen, a spatial period Tpix is ​​therefore defined on each elementary periodic grid 30 of the pseudoperiodic coding pattern 3, as being equal to the spatial period of this elementary periodic grid 30, and therefore as equal to the dimension of its patterns 301.

[0046] As illustrated in [Fig.l], the pseudoperiodic encoding target 3 is arranged on the displacement stage 14 so that the imaging instrument 1 forms an initial image Io of at least a portion of the pseudoperiodic encoding target 3 on the matrix image detector 16.

[0047] The imaged portion corresponds to the field of view of the imaging instrument 1, as defined by the matrix image detector 16 and the imaging optical system 10.

[0048] The matrix image detector 16 comprises a matrix of pixels of rectangular shape. Pixels 162 are therefore distributed according to N1 rows and M1 columns, for a total of N1xM1 pixels. The symbol x represents here, and in the remainder of the description, a multiplication between two factors, here N1 and M1, and corresponds to a notation known to those skilled in the art to indicate dimensions.

[0049] For example, the matrix image detector 16 considered in the embodiment comprises a total number NlxMl of pixels 162 equal to 400x512 pixels, or 204,800 pixels. This matrix image detector 16 corresponds for example to a CCD sensor (for Charge-Coupled device in English) or to a CMOS sensor (for Complementary metal-oxide-semiconductor, still according to the Anglo-Saxon name). The pixels 162 of the matrix image detector 16 are thus adapted to convert incident photons into electrical charges. Here, in the embodiment presented, the matrix image detector 16 corresponds to a CMOS sensor.

[0050] These electrical charges are then read on the pixels 162 of the matrix image detector 16, in order to reconstruct an image.

[0051] The matrix image detector 16 is also described by a physical size, which corresponds to a size dependent on the number of pixels 162, here, NlxMl, and the physical dimension of each of these pixels 162.

[0052] Depending on a total magnification of the imaging instrument and the physical size of the detector, a part of the pseudoperiodic encoding target 3 is imaged. As a reminder, this part corresponds to the field of view of the matrix image detector 16.

[0053] The physical size of the matrix image detector 16 is usually chosen to be large, favoring a large number of small pixels. Indeed, here, NI, the number of pixels in rows is equal to 400 pixels, and Ml, the number of pixels in columns is equal to 512 pixels, or 204,800 pixels, so as to obtain the widest possible field of view, which is advantageous for studying an observation object or observing an extended surface of the pseudoperiodic coding target 3. Thus, observations on the same observation object and / or on the pseudoperiodic coding target 3 at several magnification values ​​are facilitated by a vision on a larger surface.

[0054] The portion of the pseudoperiodic encoding pattern 3 imaged on the pixel matrix of the matrix image detector 16 is shown in [Fig.2] and [Fig.3]. Here, it is considered that the magnification of the imaging instrument 1 and the area of ​​the pseudoperiodic encoding pattern 3 included in the field of view are chosen so that at least one elementary periodic grid 30 is entirely imaged on the matrix of NlxMl pixels of the matrix image detector 16. In other words, the image of at least one elementary periodic grid 30 enters entirely in the field of view of the matrix image detector 16.

[0055] In [Fig.3], three elementary periodic grids 30 are fully imaged on the pixel matrix of the matrix image detector 16.

[0056] Preferably, it is also assumed that an image of an elementary periodic grid 30 is located close to a center of the matrix of the matrix image detector 16, in particular for reasons of image quality, and that the pattern 301 included in the elementary periodic grid 30 is perfectly resolved by the imaging instrument 1. Indeed, the image quality is better at the center of the field of view, given the reduction in optical aberrations likely to degrade the image quality.

[0057] Thus, in a space of the matrix image detector 16, it is possible to define a size in pixels of the objects, such as the pseudoperiodic coding target 3, imaged on the pixel matrix of this detector. By size in pixels, it is meant size in number of pixels.

[0058] For each of the fully imaged elementary periodic grids 30, it is possible to define at least one value of dimension 302 in pixels. This at least one dimension 302 corresponds to the length in pixels of the sides of the rectangular shape delimiting the elementary periodic grid 30. Here the elementary periodic grid 30 is defined as being delimited by a square shape, therefore a single dimension 302 in pixels is sufficient to describe the size of the elementary periodic grid 30 on the matrix image detector 16.

[0059] Furthermore, on each of the elementary periodic grids 30, as explained in upstream of the description, an elementary spatial period Tpix was defined as being equal to a dimension in pixels of a box of the elementary periodic grid 30, where a pattern 301 is arranged. In [Fig.3] in particular, Tpix corresponds to the dimension in pixels of the sum of the sides of a white square and a black square, when the pattern 301 corresponds to a checkerboard of dimension 2x2, or to the dimension in pixels of a side of a large black or white square when the pattern 301 corresponds to a plain square.

[0060] By construction, the dimension 302 in pixels of an elementary periodic grid 30 is therefore equal to an integer multiple of the elementary spatial period Tpix in pixels. This integer multiple is noted k in the remainder of the description. In the embodiment shown in [Fig.3], the elementary periodic grid 30 is divided into a grid of ten by ten patterns 301. In other words, the dimension 302 of the elementary periodic grid is here equal to ten times the elementary spatial period Tpix.

[0061] These different sizes in pixels are indicated in [Fig.3].

[0062] In this example, the elementary spatial period Tpix is ​​equal to 16 pixels, and the dimension 302 in pixels of the elementary periodic grid 30 is therefore equal to 160 pixels.

[0063] These pixel quantities are for example determined by an analysis using a processing algorithm adapted to the extraction of the orientation and / or the position of the target in order to extract therefrom, among other things, the orientation and the position of an observation object (i.e. a sample), in a manner which is mentioned in the document EP3458896 cited previously.

[0064] The controller 160 is adapted to communicate with the matrix image detector 16. Here, it is for example a computer connected to the matrix image detector 16 by an Ethernet cable, a USB cable, or any other cable corresponding to a suitable communication protocol, and left to the discretion of those skilled in the art.

[0065] More precisely, the controller 160, the matrix image detector 16 as well as the processing unit 18 are interfaced here using USB 3.0 connectors.

[0066] The control of this controller 160 is for example automated, in particular by receiving instructions from the processing unit 18, which will then be communicated to the matrix image detector 16.

[0067] Alternatively, the control of the controller 160 is carried out by a user, possibly from suggestions emanating from the processing unit 18 via a user interface.

[0068] The matrix image detector 16 is associated with acquisition parameters. These acquisition parameters, as the name suggests, are parameters whose values ​​determine the conditions for acquiring an image on the matrix image detector 16.

[0069] These acquisition parameters include in particular the exposure time, a pixel reduction parameter in B1 lines and / or B2 columns, a cropping parameter in N2 lines and / or M2 columns, as well as parameters linked to the management of parasitic noise for example. In the case of a series of several images to be acquired, an imaging rate, a number of images in the series, or even an acquisition duration are also possible acquisition parameters of the matrix image detector 16.

[0070] Generally, these acquisition parameters describe the manner in which the electrical charge value information is read on the NlxMl pixels of the pixel matrix of the matrix image detector 16, then transmitted to the processing unit 18.

[0071] The pixel reduction parameters in lines B1 and / or in columns B2 designate a pixel binning parameter in lines and / or in columns typically present on CCD or CMOS type matrix image detectors.

[0072] This parameter for grouping pixels into rows and / or columns corresponds in particular to a whole and positive numerical value.

[0073] Pixel grouping consists of forming artificial groupings of pixels of dimension BlxB2, in order to create artificial pixels, also called superpixels 164, larger in size. This has several advantages, including the reduction of the reading time on the matrix image detector 16, because the number of pixels in the acquired image is reduced by the factor B lxB2.

[0074] The pixels 162 or superpixels 164 being considered in the same way by the matrix image detector 16, that is to say as a single pixel where an accumulated charge value must be accumulated and read. Thus, the dimensions described in pixels in the remainder of the description can refer both to dimensions in pixels 162, or in superpixels 164.

[0075] Usually, the pixels 162 are grouped into groups so as to form artificial square pixels, that is to say that the pixel reduction parameter in rows B1 has the same numerical value as the pixel reduction parameter in columns B2.

[0076] For example, in the case of a grouping of pixels called "2x2", a set of two adjacent pixels in rows and two superimposed pixels in columns, i.e. a square of four pixels 162, are read as a single superpixel 164 whose intensity is equal to the average of the intensity of the four pixels. Likewise for a grouping of pixels called "4x4", where a set of four adjacent pixels in rows and four superimposed pixels in columns, forming a square of 16 pixels 162, are read as a single superpixel 164 whose intensity is equal to the average of the intensity of the 16 pixels 162.

[0077] According to a possible variant, the pixel reduction parameters in rows B1 and / or in columns B2 designate a pixel subsampling parameter in rows and / or columns. In this case, the electric charge value is only read on certain pixels, all the B1 pixels along a row and all the B2 pixels along a column.

[0078] In the remainder of the description, it is considered that the pixel reduction parameters in rows B1 and / or in columns B2 designate pixel grouping parameters in rows and / or in columns. The transposition of the content of the present disclosure to the case of a sub-sampling parameter in rows and / or in columns is however within the reach of those skilled in the art.

[0079] The row cropping parameter N2 and / or column cropping parameter M2 delimits on the pixel matrix of the matrix image detector 16 a region of interest in pixels (often referred to as ROI, from the English term “Region of Interest”). This region of interest in pixels extends over N2 pixels (or superpixels, if a different parameter for grouping pixels in rows and / or columns equal to one is applied) adjacent on a row of the pixel matrix, and similarly over M2 pixels (or superpixels) adjacent in columns. The number N2 is an integer less than or equal to NI, and, similarly, the number M2 is an integer less than or equal to ML

[0080] As for the pixel reduction parameters in rows B1 and / or in columns B2, the cropping parameters in rows N2 and / or in columns M2 correspond in particular to a whole and positive numerical value.

[0081] These acquisition parameters have default values, usually applied automatically to the matrix image detector 16. Thus, by default, the number of pixels of the matrix image detector 16 read during the acquisition of an image is not reduced by pixel grouping or by pixel subsampling, and is by default NlxMl.

[0082] For example, in the case of pixel grouping, the row pixel reduction parameter B1 is equal to one by default, and the column pixel reduction parameter B2 is also equal to one. This example is illustrated in [Fig.3].

[0083] In other words, the electric charge value information is read individually on each of the NlxMl pixels physically making up the pixel matrix of the matrix image detector 16.

[0084] Similarly, the row cropping parameter N2 and the column cropping parameter M2 also have default values, automatically applied by default on the matrix image detector 16. All the physical pixels in the extent of the pixel matrix are read, so the row cropping parameter N2 is equal to the number NI of physical pixels. The same applies to the column cropping parameter M2, which is then equal to the number M1 of physical pixels corresponding to the number of columns.

[0085] An image of a part of the pseudoperiodic coding pattern 3, comprising at least in its entirety an elementary periodic grid 30 acquired by the matrix image detector 16 when the default acquisition parameters are applied is designated by the term initial image Io, this initial image Io corresponding to the image represented in [Fig.3].

[0086] Thus, this initial image Io has a dimension of NlxMl pixels. It is therefore a full-field image, since all the NlxMl pixels of the pixel matrix are read, and an image with full spatial resolution, because the pixels are neither grouped nor sub-sampled.

[0087] This initial image Io acquired by the matrix image detector 16 where the controller has applied the default acquisition parameters, is then analyzed using suitable processing algorithms, in order to extract therefrom, among other things, the orientation and / or the position of a sample and / or of the pseudoperiodic coding target 3.

[0088] For this, it is provided that the matrix image detector 16 communicates with the processing unit 18. Indeed, it is considered here that the processing unit 18 is configured to implement the processing algorithm adapted to the extraction of information from the image of the pseudoperiodic coding test pattern 3.

[0089] Thus, the matrix image detector 16 transfers the initial image Io to the processing unit 18, the initial image Io having a weight of 1 byte per pixel (or superpixel), i.e. a total weight of 204,800 bytes. As a reminder, one byte, or one byte corresponding to eight bits.

[0090] The weight in bytes per pixel is indicated here for the case of a grayscale image coded over a dynamic range of 256 distinct values. Heavier images, in particular color images, or images on grayscale associated with wider ranges of values ​​are also possible, but are not of particular interest in the context of extracting information from a pseudoperiodic 3-coded test pattern image.

[0091] The processing unit 18 mentioned here is for example part of a computer. The controller 160 and the processing unit 18 are possibly integrated into the same computer, but nevertheless play distinct roles within the framework of the present disclosure. Alternatively, the controller 160 and the processing unit 18 are integrated into two physically distinct elements, for example two computers, or the controller 160 is integrated into a separate module provided with a user interface.

[0092] This processing unit 18 is designed to apply the processing algorithms adapted for the extraction of information from the pseudoperiodic coding pattern.

[0093] [Fig.4] shows an image called a reduced image L of the pseudoperiodic coding target 3, after application of a method for acquiring images of the target pseudoperiodic coding for micro-location being the subject of the present disclosure.

[0094] The invention in fact advantageously proposes to determine acquisition parameters, called reduced acquisition parameters, making it possible to reduce the acquisition time per image and, by extension, to increase the imaging rate, also called acquisition frequency. This also has the consequence of reducing the processing time by the processing algorithm adapted to the extraction of information from the image of the pseudoperiodic coding test pattern 3. This determination of the reduced acquisition parameters is done while retaining in particular a sufficient quantity of spatial information in the reduced image h for the proper functioning of the processing algorithm adapted to the extraction of information from the image of the pseudoperiodic coding test pattern 3.

[0095] [Fig.5] represents such a process in the form of a flowchart.

[0096] A first step consists of the acquisition E0 of an initial image Io by the matrix image detector 16, as described previously. As a reminder, for the recording of this image, the controller 160 applies the default acquisition parameters to the matrix image detector 16. The initial image Io therefore corresponds to a full-field image, and at full spatial resolution.

[0097] The processing algorithm adapted to the extraction of information from the image of the pseudoperiodic coding pattern is applied to this initial image Io.

[0098] The second step of the method consists of determining El of the elementary spatial period Tpix in pixels of the elementary periodic grid 30 from the pseudoperiodic coding pattern 3, this quantity Tpix and its definition having already been described previously in the present disclosure.

[0099] This determination El of the elementary spatial period Tpix in pixels is advantageously made from the initial image Io acquired in the first step, by applying thereto the processing algorithm adapted to the extraction of information from the image of the pseudoperiodic coding test pattern. Thus, the information of the value in pixels of the elementary spatial period Tpix is ​​for example part of the information obtained at the end of the processing algorithm, or of an analysis of the image by spatial Fourier transform. It is in particular particularly advantageous to extract the value of the elementary spatial period Tpix via an analysis of the initial image Io by spatial Fourier transform, because this method allows an extraction of a value with a precision lower than the pixel.

[0100] Alternatively, an image processing algorithm, via shape recognition, can be applied to the initial image Lafin to extract the pixel value of the elementary spatial period Tpix.

[0101] According to another variant, this elementary spatial period value T is determined pix into pixels via a storage unit, not shown here, or from an estimate of its value based on experiment parameters. Indeed, since this elementary spatial period value Tpix is ​​related to the physical size of the pattern 301 on the pseudoperiodic coding target 3, to the magnification of the imaging instrument 1 and to the physical size of the pixels of the matrix image detector 16, a rough estimate can be established and stored in memory.

[0102] For example, the elementary spatial period Tpix is ​​determined by the processing algorithm adapted to the extraction of information from the image of the pseudoperiodic coding test pattern 3 as being equal to 16 pixels.

[0103] The third step consists of calculating E2 the pixel reduction parameters in rows B1 and / or in columns B2. Here, for example, the pixel reduction parameter in rows B1 is equal to the reduction parameter in columns B2.

[0104] Furthermore, this calculation E2 is carried out by the processing unit 18.

[0105] The method of calculating the pixel reduction parameters in B1 lines and / or in B2 columns allowing to maintain a spatial resolution necessary to resolve a pattern 301 of pixel dimension Tpix is ​​detailed in [Fig.6].

[0106] It is first considered that the row reduction parameter B1 is equal to the column reduction parameter B2, and that it is a reduction by grouping. In the case of an absence of grouping, these two parameters are worth one, this is the default value of these parameters.

[0107] According to a variant, the calculation step E2 is based on a sub-sampling of the image by the detector.

[0108] It is considered that the processing unit 18 accesses a storage unit in the sub-step E21, where a minimum value of spatial period in pixels, called minimum spatial period, is stored. This value reflects the minimum spatial resolution in pixels necessary for the pattern 301 to be able to be interpreted correctly and unambiguously by the imaging instrument 1. In other words, here, if the elementary spatial period Tpixen pixel is less than the minimum spatial period, in particular if the pattern 301 considered is a checkerboard of 2x2 black and white squares, this pattern 301 is then imaged by the imaging instrument 1 as a gray area, which may possibly be confused with a plain pattern 301. In which case, the information necessary for the proper functioning of the processing algorithm adapted for obtaining information from the image of the pseudoperiodic encoding test pattern is lost, degrading the performance of this algorithm.

[0109] This minimum spatial period is linked, among other things, to the resolving power, also called spatial resolution, of the optical imaging system 10, i.e. its ability to distinguish spatially close details. It is considered that this value is known upstream of the process, and stored. It is also possible to determine it from of an image analysis, according to techniques known to those skilled in the art, for example by calculating a derivative of the image of an edge, acquired in full field, at maximum spatial resolution. By maximum resolution, it is understood that all the pixels are taken into account, and that the pixel reduction parameters in rows and / or columns are equal to one, corresponding to a case where the pixels are not grouped

[0110] Thus, it is essential that even after reduction of the pixels, thanks to a grouping in order to form superpixels for example, the new elementary spatial period Tpix in pixels remains greater than the minimum spatial period in pixels.

[0111] In the case where the elementary spatial period Tpix is ​​greater than or equal to twice the minimum spatial period, then a new value for the row reduction parameter B1 and / or column reduction parameter B2 can be calculated, in substep E23. In this case, B1 = B2 = ' oa 'a function int() designates the function "part integer”, which returns an integer value, truncated of the decimals and not rounded, and where Tmin denotes the minimum spatial period in pixels. This condition is translated by substep E22 of the flowchart in [Fig.6].

[0112] Here, the value of the minimum spatial period is equal to 5 pixels, and the value of the elementary spatial period is initially equal to 16 pixels, as mentioned previously.

[0113] A new elementary spatial period value Tpix is ​​also calculated in sub-step E24. Thus, Tpix = Tpix / B1; in the case where the elementary periodic grid 30 is square in shape, and B1 and B2 are equal.

[0114] If the condition is not satisfied, then the value of the pixel reduction parameter in rows B1 and / or in columns B2 is not recalculated, which corresponds to sub-step E25.

[0115] After this calculation step E2 carried out by the processing unit 18, the latter returns at least two parameters, a value of the pixel reduction parameter in rows B1 and / or in columns B2, as well as the elementary spatial period Tpix in pixels of the elementary periodic grid 30 once the possible pixel reduction has been applied.

[0116] Here, the processing unit 18 determines a pixel reduction parameter, more precisely, a pixel grouping parameter in rows and columns equal to three.

[0117] The fourth step consists of an adjustment E3 by the controller 160 of the default acquisition parameters on the matrix image detector 16 in order to obtain a set of reduced acquisition parameters. First, the controller 160 receives from the processing unit 18 the value(s) calculated for the pixel reduction parameters in rows B1 and / or in columns B2. Then, the controller 160 communicates with the matrix image detector 16, via an appropriate communication channel, for example via an Ethernet or USB cable, in order to adjust the value(s) of the pa(s) acquisition parameters corresponding to the pixel reduction parameters in rows B1 and / or in columns B2.

[0118] This adjustment consists of replacing a previously assigned value, or the default value, with the new calculated acquisition parameter value.

[0119] It is also conceivable that the matrix image detector 16 comprises a record of at least one pre-established list of predetermined values ​​of the pixel reduction parameter in rows B1 and / or columns B2. These predetermined values ​​correspond to values ​​authorized by manufacturer parameters of the matrix image detector 16, such as, for example, integer values ​​that are multiples of two. In this case, where only certain predetermined values ​​are authorized, the controller 160 adjusts the value(s) of the acquisition parameter to the predetermined lower value closest to the value calculated for the pixel reduction parameter in rows B1 and / or columns B2. This is done in a selection sub-step E31 included in the adjustment step E3.

[0120] In the event that the values ​​of the pixel reduction parameters in rows B1 and / or in columns B2 are equal to their default values, the adjustment step E3 is optional.

[0121] Here, for example, in the matrix image detector 16 used only admits even pixel grouping values. Thus, during the adjustment E3, the controller 160 adjusts the value of the associated acquisition parameter, here the pixel reduction parameter in rows B1, and in columns B2 to the permitted value closest to and lower than the value calculated during the step E2.

[0122] Thus, the pixel reduction parameter in rows and columns actually applied to the matrix image detector 16 is equal to two in rows and columns instead of three. This pixel reduction parameter in rows B1 and / or in columns B2 of two is therefore illustrated in [Fig.4].

[0123] Secondly, at least one trimming parameter in rows N2 and / or in columns M2 is determined.

[0124] For this, a fifth step consists of the determination E10 of at least one dimension 302 in pixels of an elementary periodic grid 30 of the pseudoperiodic coding pattern 3.

[0125] In practice, the elementary periodic grid 30 being here of square shape, a single dimension 302 in pixels is sufficient to describe the size of this elementary periodic grid 30. It is also considered in this example that this dimension 302 in pixels is equal to an integer multiple, noted k, of the elementary spatial period Tpix as explained previously.

[0126] For example, the value of this integer multiple k is chosen and fixed during the design of the pseudoperiodic coding pattern 3, and knowledge of its value is integrated into software configured to operate the processing algorithm adapted for obtaining information from the image of the pseudoperiodic encoding test pattern 3.

[0127] Its value is therefore stored in the storage unit, not shown here.

[0128] Thus, the determination E10 of the dimension 302 in pixels of the elementary periodic grid 30 is executed by the processing unit 18, which here determines the value of the integer multiple k by accessing the storage unit.

[0129] The value of the dimension 302 in pixels is then deduced from the value of the determined integer multiple k and the value Tpix.

[0130] Other methods of determining the integer value k are left to the discretion of those skilled in the art.

[0131] It should also be noted that the value Tpix may correspond to a value in pixels possibly after reduction of the number of pixels by agglomeration into superpixels. In other words, the pixel reduction parameters in rows B1 and / or in columns B2 have already been calculated, and a new possible value of Tpix could have been returned by the processing unit 18. The value of the integer multiple k is however not modified during the method.

[0132] Here, the integer multiple k determined during step E10 is equal to ten.

[0133] A sixth step then consists of the calculation E20 of at least one cropping parameter in rows N2 and / or in columns M2 as a function of the dimension 302 in pixels determined for the elementary periodic grid. In practice, it is the value of the pre-recorded integer multiple k and the value of the elementary spatial period Tpix which are used during this calculation.

[0134] Typically, this calculation step E20 is executed by the processing unit 18.

[0135] In the most general case, it is assumed that the sides of the periodic grid ele mentary 30 are not approximately aligned with the two principal axes of the pixel array of the matrix image detector 16 or that the orientation of the two principal axes of the pixel array relative to the sides of the elementary periodic grid 30, defining an angle 0 is unknown. This angle 8 is shown in [Fig.3]

[0136] In this case, an optimal image dimension, in order to have at least one fully imaged elementary periodic grid, is determined by the following formula, always in the case where the elementary periodic grid 30 is square in shape: N2 = M2 = Cl * k * TPiX; where k corresponds to the integer multiple, so that k * Tpix is ​​equal to the dimension 302 in pixels of the elementary periodic grid 30, Tpix is ​​equal to the elementary spatial period in pixels, possibly updated by the method. In which case, the value of Tpix corresponds to superpixels 164, assimilated to pixels 162. Cl is equal to a calculation constant, dependent on the algorithmic treatments used, the symbol * designates a multiplication and where the symbol V corresponds to the square root function.

[0137] Here, Cl is equal to two.

[0138] Thus, a cropping parameter N2 in rows and a cropping parameter M2 in columns are determined, which correspond to the dimension in pixels of the image after cropping.

[0139] In the case where the orientation of the two main axes of the pixel matrix relative to the sides of the elementary periodic grid 30, defining an angle 0 is known, for example, thanks to the information from the processing algorithm, adapted to the extraction of information, in position and / or in orientation.

[0140] In the case where the angle 0 is less than 5 degrees, which corresponds to a situation where the person skilled in the art, as well as an operator or experimenter, judges that the sides of the elementary periodic grid 30 are approximately aligned with the two main axes of the pixel matrix of the matrix image detector 16. Then, still following the aforementioned conditions N2 = M2 = Cl * k * Tpix; where the symbols are the same as previously.

[0141] Here, in the example, the angle 6 with respect to the axes of the matrix, being equal to 1.6 degrees, the formula is applicable.

[0142] Thus, the formulas proposed for the calculation E20 of the cropping parameter in lines N2 and / or in columns M2 offer a compromise between a minimum image size, and the need to include in full at least one elementary periodic grid 30.

[0143] Here, the processing unit 18 calculates a value for the row cropping parameter N2 equal to 160 pixels, and a column cropping parameter M2 equal to 160 pixels. In practice, in the embodiment described, given that the pixel reduction parameter in rows B1 and / or in columns B2 is different from one, these pixels actually correspond to superpixels. Since they are equivalent with respect to the matrix image detector 16, the two notions are merged here.

[0144] Once this calculation E20 has been carried out, the determined value(s) are communicated to the controller 160. The latter, during a seventh step, carries out an adjustment E30 of at least one acquisition parameter of the matrix image detector 16, thanks to the at least one image cropping parameter in lines N2 and / or columns M2 thus determined, in order to obtain reduced acquisition parameters.

[0145] In particular, according to a possible variant, the authorized values ​​for the cropping parameters in rows N2 and / or in columns M2 depend on the pixel reduction parameters in rows B1 and / or in columns B2. For example, it may be imposed that the values ​​for the cropping parameters in rows N2 and / or in columns M2 correspond to integer multiples of the pixel reduction parameter in rows B1 and / or in columns B2. or in B2 columns. This avoids obtaining isolated pixels, which are not grouped, or irregular subsampling.

[0146] During an eighth step, the matrix image detector 16 acquires a new image, called reduced image Ib, by applying the reduced acquisition parameters determined in the previous steps.

[0147] A center of the cropped image extends from a center of the initial field, defined on the NlxMl pixels of the pixel matrix. Indeed, the central part of the field being less degraded by aberrations, it is advantageous to keep it in the cropped image of dimension N2xM2 pixels (or superpixels, in a perfectly equivalent manner).

[0148] Nevertheless, a cropped image of dimension N2xM2 pixels (or superpixels) contained on any region of the matrix of NlxMl pixels is also conceivable within the framework of the present disclosure.

[0149] This reduced image Ib is possibly of a lower spatial resolution compared to the initial image Io, due to the application of pixel reduction parameters in lines B2 and / or in columns Bl, and possibly of smaller size in number of pixels due to the application of a cropping parameter in lines N2 and / or in columns M2.

[0150] Here, as a reminder, the reduced image ^represented in [Fig.4] is acquired by having applied to the matrix image detector 16 a pixel grouping parameter in rows and columns equal to two, and a cropped image dimension of 160 pixels by 160 pixels.

[0151] Thus, in general, the reduced image I thus obtained is less heavy than the initial image Io, read over the entirety of the NI lines and Ml columns of pixels, each of these pixels being taken into account, and read.

[0152] Indeed, the total weight of the reduced image h is reduced by a factor of eight compared to the initial image Io.

[0153] The reading time, or acquisition time, which is limited by a transfer rate defined on the matrix image detector 16 (in bytes per second, or bytes per second) depends on the total weight, and therefore on the size of the reduced image Ib. Consequently, the reduction in the total weight of the reduced image h carried out using the method which is the subject of the present disclosure, results in a reduction in the acquisition time.

[0154] In the case where a series of images is acquired, i.e. a film, the reduction in the acquisition time also advantageously implies an increase in the imaging rate. This imaging rate here corresponds to a frequency at which a new image can be acquired on the matrix image detector 16.

[0155] This is particularly advantageous when, as in the application described in the preamble, a mechanical stability study is carried out, since the increase in the frame rate makes it possible to observe transient phenomena.

[0156] Thus, in the case of an initial image Io acquired using the default acquisition parameters, the maximum imaging rate is limited to 50 images per second, while the use of the reduced acquisition parameters determined according to the method described makes it possible to increase the imaging rate by a factor of eight. This is therefore equal to 400 images per second.

[0157] Similarly, the processing time required by the processing algorithm adapted to extract a position from the reduced image h goes from 0.5 seconds to 0.06 seconds.

[0158] The application of the method described to a matrix image detector model 16 different from the model described so far is also possible. For example, a matrix image detector 16 is considered having NI equal to 1920 pixels in rows and Ml equal to 2560 pixels in columns. An analysis of an initial image Io acquired by this matrix image detector 16, using the default acquisition parameters, makes it possible to determine that the elementary spatial period Tpix is ​​here equal to 80 pixels, while the minimum spatial value is equal to 11 pixels. Using this data, it is calculated by the processing unit 18 that the pixel reduction parameter in rows B1 and the pixel reduction parameter in columns B2 are equal to 7. However, as for the detector considered previously, the matrix image detector 16 only accepts even values ​​for these parameters.Thus, the controller 18 adjusts the acquisition parameters of the matrix image detector 16 to the nearest lower even value. The row pixel reduction parameter B1 is therefore equal to 6, as is the column pixel reduction parameter B2, which is also equal to 6.

[0159] Still according to the method, a row cropping parameter N2 and a column cropping parameter M2 are also determined. For this, a dimension 302 in pixels of the elementary grid 30 is determined as being equal to an integer multiple of the elementary spatial period Tpix, here, after redefinition of its value during the calculation of the pixel reduction parameter in rows B1 and in columns B2. As previously, since the same pseudoperiodic coding pattern 3 is considered, this integer multiple, noted k, is equal to 10. The processing unit 18 thus calculates a value for the row cropping parameter N2 and in columns M2 which is equal to 260 pixels.

[0160] Thus, thanks to the method, the number of pixels in the reduced image h increases to 136,640 pixels, compared to 4,915,200 pixels in the initial image Io, i.e. a factor of seventy-three between the two images.

[0161] Here, the maximum acquisition rate increases from 10 images per second for the default acquisition parameters to 500 images per second for the reduced acquisition parameters, i.e. a factor of fifty. Note that the acquisition rate is here limited no longer by the reading time of the charges accumulated on the pixels 162 or superpixels 164 of the detector matrix, but by incompressible times.

[0162] The processing time necessary for the processing algorithm adapted to extract a position from the reduced image h nevertheless drops from 5 seconds to 0.07 seconds.

[0163] Other embodiments of the method described above are also possible. In particular, according to one of these embodiments, in a first step, a first reduced image h is acquired by applying pixel reduction parameters in rows B1 and / or in columns B2 different from the default parameters. In a second step, cropping parameters in rows N2 and / or in columns M2 are determined. Here, the dimension 302 in pixels of the elementary cell 30 is determined from the elementary spatial period Tpix and the value of k, or from an analysis of the reduced image Ii acquired upstream, and for which the value of the elementary spatial period Tpix has already been updated when this proves necessary.

[0164] According to other embodiments, a first image is acquired using arbitrary reduced acquisition parameters, for example with a pixel reduction parameter in rows B1 and / or in columns B2, corresponding to a grouping parameter, equal to two in rows and columns for example, i.e. B1=B2=2, before performing an analysis thereof. In the case where the spatial resolution is not sufficient for an analysis by the processing algorithm adapted for the extraction of information from the image of the pseudoperiodic coding test pattern 3, the method is then applied.

[0165] Other embodiments, with various default acquisition parameters are within the reach of those skilled in the art.

[0166] The present invention is in no way limited to the embodiments described and shown, but those skilled in the art will be able to provide any variant in accordance with the invention.

Claims

Claims

1. Method for acquiring images of a pseudoperiodic coding pattern (3) for micro-localization using a matrix image detector (16), comprising N1 rows and M1 columns of pixels, the matrix image detector (16) being included in an imaging instrument (1), the matrix image detector (16) being controlled by a controller (160) connected to a processing unit (18), and having acquisition parameters, including at least one parameter for reducing pixels into rows (B1) and / or columns (B2), the pseudoperiodic coding pattern (3) comprising an elementary periodic grid (30) in which at least one pattern (301) is arranged, said method being characterized in that it comprises the following steps: A) determination (El) by the processing unit (18) of an elementary spatial period (Tpix) in pixels of the elementary periodic grid (30) from a first image of the pattern acquired by the detector of matrix images (16),B) calculation (E2) by the processing unit (18) of at least one pixel reduction parameter in rows (B1) and / or in columns (B2) as a function of the elementary spatial period (Tpix) in pixels and of a minimum spatial period in pixels, C) adjustment (E3) of at least one acquisition parameter of the matrix image detector (16) by the controller (160) using the at least one pixel reduction parameter in rows (B1) and / or in columns (B2) transmitted by the processing unit (18), to obtain reduced acquisition parameters, D) acquisition (E4) of a reduced image (R) of the pseudoperiodic coding test pattern (3) by applying the reduced acquisition parameters to the matrix image detector (16).,

2. Method according to claim 1, in which the at least one parameter for reducing pixels into rows (B1) and / or columns (B2) comprises a parameter for grouping pixels into rows and / or columns.

3. Method according to claim 1, wherein the at least one parameter for reducing pixels in rows (B1) and / or in columns (B2) comprises a parameter for subsampling pixels in rows and / or in columns.

4. A method according to any one of claims 1 to 3, wherein the matrix image detector (16) comprises a recording of at least at least one pre-established list of predetermined values of the pixel reduction parameter in rows (B1) and / or in columns (B2), the method comprising during step C) of adjustment (E3) of at least one acquisition parameter (Param), a sub-step of selection (E31) of at least one value in the recording according to the calculation (E2) of step B).

5. Method according to any one of claims 1 to 4, in which during step A), the first image, corresponding to an initial image (Io), is read individually on the entirety of the NI lines and Ml columns of pixels (162) of the matrix image detector (16), taken one by one

6. 1111. Method according to any one of claims 1 to 5, in which the following steps are added between step B) of determining (El) an elementary spatial period (Tpix) of the elementary periodic grid (30) and step D) of acquiring (E4): E) determining (E10) by the processing unit (18) at least one dimension (302) in pixels of an elementary periodic grid (30) originating from the pseudoperiodic coding pattern (3), F) calculating (E20) by the processing unit (18) at least one cropping parameter in rows (N2) and / or in columns (M2) as a function of at least one dimension (302) in pixels of an elementary periodic grid (30) originating from the pseudoperiodic coding pattern (3), G) adjusting (E30) at least one acquisition parameter of the matrix image detector (16) by the controller (160) with the at least one row (N2) and / or column (N2) and / or column (M2) trimming parameter transmitted by the processing unit (18),to obtain reduced acquisition parameters.,

7. Method according to claim 6 wherein, during step E), the dimension (302) in pixels of the elementary grid (30) is determined from the value of an integer multiple stored in a storage unit and the elementary spatial period (Tpix) in pixels of the elementary periodic grid (30).

8. Method according to claim 6 wherein, during step E), the dimension (302) in pixels of the elementary grid (30) is determined from an image acquired on the matrix image detector (16).

9. Method according to any one of claims 6 to 8, in which step F) of calculating (E20) at least one parameter for trimming in lines (N2) and / or in columns (M2) takes into account an orientation (0) of the elementary periodic grid (30) with respect to two axes of the matrix image detector (16).

10. Method according to any one of claims 6 to 9, in which the at least one value calculated in step F) for the at least one row (N2) and / or column (M2) cropping parameter takes into account the row (B1) and / or column (B2) pixel reduction parameters calculated in step B).

11. Method according to any one of claims 1 to 10, in which the method is adapted so that the acquisition step D) (E4) requires an acquisition time less than an acquisition time necessary for an initial image acquisition (Io), the initial image (Io) being read on the entirety of the NI lines and Ml columns of pixels (162) of the matrix image detector (16), taken one by one.

12. Device for acquiring images of a pseudoperiodic coding pattern (3) for micro-localization comprising a processing unit (18) and a matrix image detector (16) comprising N1 rows and M1 columns of pixels (162), and having acquisition parameters including pixel reduction parameters in rows (B1) and / or in columns (B2), the matrix image detector (16) being controlled by a controller (160) connected to the processing unit (18), said device being characterized in that: - the processing unit (18) is adapted to determine an elementary spatial period (Tpix) in pixels of an elementary periodic grid (30) contained in the pseudoperiodic coding pattern (3) from a first image of the pattern acquired by the matrix image detector (16),and to calculate at least one pixel reduction parameter in rows (Bl) and / or in columns (B2) as a function of the elementary spatial period (Tpix) and a minimum spatial period in pixels, - the controller (160) being adapted to receive at least one pixel reduction parameter in rows (Bl) and / or in columns (B2) from the processing unit (18), - the controller (160) being adapted to adjust at least one acquisition parameter of the matrix image detector (16) as a function of the at least one pixel reduction parameter in rows (Bl) and / or in columns (B2), to obtain reduced acquisition parameters, - the matrix image detector (16) being adapted to acquire a reduced image (Ii) of the pseudoperiodic coding test pattern (3) by applying the reduced acquisition parameters.,

Images