METHOD FOR OBSERVING A SAMPLE BY LENSLESS IMAGING, TAKING INTO ACCOUNT SPATIAL DISPERSION WITHIN THE SAMPLE

Abstract

Method for observing a sample (10) comprising the following steps: a) illumination of the sample using a light source (11), the light source emitting an incident light wave (12) propagating towards the sample (10) along a propagation axis (Z); b) acquisition, using an image sensor (16), of an image (I0) of the sample (10), formed in a detection plane (P0); c) formation of a stack of images, called reconstructed images, from the image acquired (I) in step b), each reconstructed image being obtained by applying a digital propagation operator (h), according to a reconstruction distance (zj); d) from each image (Azj) of the image stack, calculation of a sharpness indicator (Cj(x,y)) for different radial positions (x,y), each sharpness indicator being associated with a radial position (x,y) and a reconstruction distance (zj).

Description

TECHNICAL FIELD The technical field of the invention is the observation of a sample, in particular a sample comprising particles, by lensless imaging, or by so-called defocused imaging. PREVIOUS ART Lensless imaging has developed in recent years, and its applications for sample characterization are multiplying, for example in the fields of medical diagnostics, cell culture, biological analysis or in the food industry. This technique makes it possible to obtain, at low cost, an image of a sample with a high field of observation. The application of holographic reconstruction algorithms to images obtained using a lensless configuration makes it possible to obtain usable images of samples, which constitutes a low-cost alternative to conventional microscopy, while benefiting from a wide field of observation. Various documents describe the use of lensless imaging for observing samples. For example, document WO2016189257 describes the application of lensless imaging, coupled with holographic reconstruction algorithms, to the observation of tissue slides for diagnostic purposes. Documents WO2016151248 or WO2017178723 describe the use of lensless imaging to characterize cells in a sample. The technique described in these documents is based on the acquisition of an image using an image sensor placed in front of a sample. The application of a holographic reconstruction algorithm makes it possible to form profiles representative of the light wave to which the image sensor is exposed. From the profiles, the cells placed in the sample are characterized.The use of profiles allows for insensitivity to the position of a cell to be characterized in the sample, and in particular to the distance between the cell and the sample. A difficulty associated with the application of holographic reconstruction algorithms is that it is necessary to know the distance between the image sensor and the sample. However, when the sample contains particles, the position of the latter, relative to the image sensor, can vary within the thickness of the sample. Thus, images obtained by holographic reconstruction can be sharp for some particles, but blurred for other particles. The invention overcomes this difficulty by enabling the formation of an image in which different particles, located at different distances from the image sensor, appear sharp, without requiring multiple image acquisitions. Thus, the invention makes it possible to increase the depth of field of lensless imaging. STATEMENT OF THE INVENTION A first object of the invention is a method for observing a sample comprising the following steps: (a) illumination of the sample using a light source, the light source emitting an incident light wave propagating towards the sample along a propagation axis; b) acquisition, using an image sensor, of an image of the sample, formed in a detection plane, the sample being arranged between the light source and the image sensor, each image being representative of an exposure light wave, to which the image sensor is exposed under the effect of the illumination, the image comprising pixels, each pixel corresponding to a radial position defined in a plane parallel to the detection plane; c) forming a stack of complex images, called reconstructed images, from the image acquired during step b), each reconstructed image being obtained by applying a digital propagation operator, according to a reconstruction distance, along the propagation axis, the stack of images comprising as many reconstructed images as there are different reconstruction distances, each reconstructed image being representative of an exposure light wave to which the image sensor is exposed; d) from each image of the image stack, calculating a sharpness indicator for different radial positions, each sharpness indicator being associated with a radial position and a reconstruction distance; the method being characterized in that it also comprises the following steps: (e) taking into account a selection criterion; (f) for each radial position, and depending on the selection criterion, selection of a sharpness indicator, from among the different sharpness indicators defined, at the radial position, from the different reconstructed images, the selected sharpness indicator being associated with an optimal reconstructed image for the radial position considered; g) forming an observation image of the sample, each pixel of which is associated with a radial position, the pixel value of the observation image, or even of each pixel of the observation image, being determined as a function of the value, at the radial position of the pixel, of the optimal reconstructed image. The method may comprise, following step f), a determination, for each radial position considered, of an optimal distance, corresponding to the reconstruction distance of the optimal reconstructed image associated with the radial position. According to one embodiment, each sharpness indicator is calculated from an elementary image, the elementary image being established from the modulus and / or the argument and / or the real part and / or the imaginary part of a reconstructed image. An elementary image corresponds to an image formed from scalar values, obtained from a reconstructed image. Step d) may comprise, for each radial position considered, a calculation of a gradient of the elementary image, in the radial position, according to a gradient direction. According to one embodiment, step d) comprises, for each reconstructed image of the image stack: a calculation of a gradient image, representative of a gradient of the reconstructed image according to at least one gradient direction; a calculation of a norm of the gradient image; such that the sharpness indicator, at each radial position, is obtained from the norm of the gradient image at the radial position. Step f) may include taking into account a validity range, and rejecting a selected sharpness indicator if the reconstruction distance associated with it is not included in the validity range. In step g), the observation image, at each radial position, can be obtained from the optimal reconstructed image for the radial position considered. According to one embodiment, no imaging optics are disposed between the sample and the image sensor. According to another embodiment, an optical system is disposed between the sample and the image sensor, the optical system combining an image plane with an object plane, the detection plane being distant from the image plane, or the sample being distant from the object plane, the detection plane being offset from the image plane, and / or the sample being offset from the object plane, such that the image acquired by the image sensor is a defocused image of the sample. The sample may be held in or on a holding member, the sample being stationary relative to the holding member. The sample may comprise particles. The method may then comprise a step h) of characterizing the particles from the observation image of the sample, the characterization comprising: a particle count; and / or a determination of a particle size and / or shape; and / or a counting of particles according to their size and / or shape; and / or a determination of a three-dimensional position of the particles. The particles may be immobile relative to the holding element. A second object of the invention is a device for identifying a sample, comprising: - a light source configured to emit an incident light wave propagating towards the sample; - an image sensor; - a support, configured to hold the sample between the light source and the image sensor; a processor, configured to receive an image of the sample acquired by the image sensor and to implement steps c) to g) of a method according to the first subject of the invention. According to one embodiment, no imaging optics are disposed between the sample and the image sensor. According to one embodiment, the device comprises an optical system defining a focusing configuration, according to which the image sensor is conjugated with a plane passing through the sample, the device being such that the image sensor or the sample are offset relative to the focusing configuration, such that during step b), the image sensor acquires a defocused image of the sample. Other advantages and characteristics will emerge more clearly from the following description of particular embodiments of the invention, given as non-limiting examples, and represented in the figures listed below. FIGURES Figure 1A represents an example of a device according to the invention. Figure 1B represents a light source in one embodiment of the invention. Figure 2A shows an image of a sample acquired by an image sensor. Figure 2B shows an image reconstructed according to the prior art, considering an "average" reconstruction distance. Figure 2C shows an image referred to as an "observation image", obtained by implementing the invention. Figure 2D represents images, called elementary images, obtained on the basis of complex images reconstructed at different reconstruction distances. The elementary images form a stack of elementary images of the sample. Figure 3 illustrates the main steps of a method according to the invention. Figures 4A and 4B are images of the same sample obtained using a microscope, respectively according to two different focal planes. Figures 4C and 4D are images of the sample shown in Figures 4A and 4B, these images being obtained by lensless imaging respectively according to the prior art and by implementing the invention. Figures 5A and 5B are images of a sample obtained by lensless imaging respectively according to the prior art and by implementing the invention. Figure 6 is another example of a device suitable for implementing the invention. PRESENTATION OF SPECIAL EMBODIMENTS Figure 1A represents an example of a device according to the invention. A light source 11 is configured to emit a light wave 12, called an incident light wave, propagating towards a sample 10, along a propagation axis Z. The light wave 12 is emitted according to an illumination spectral band AX. Sample 10 is a sample that we wish to characterize. The sample comprises a medium 10m, for example a liquid medium, in which 10p particles are bathed. The particles can be cells, or microorganisms, or microalgae, or their fragment. By microorganism, we mean in particular a yeast, a bacterium, a spore, a fungus. The term particles can also designate solid particles, in particular microbeads, for example metal microbeads, glass microbeads or organic microbeads, commonly used in biological protocols. They can also be insoluble droplets bathed in a liquid medium, for example lipid droplets in an oil-in-water emulsion. The particles can have a diameter or a diagonal less than 100 pm. They can be inscribed in a circle or a sphere whose diameter is less than 100 pm. The medium 10m may be a solid medium or a gel-forming medium. The medium 10m may comprise a bodily fluid, for example, and without limitation, blood, urine, lymph or cerebrospinal fluid. It can be a culture medium, conducive to the development of cells or microorganisms. The sample 10 is held by a holding element 15. The holding element has the function of holding the sample facing the image sensor 16. It is preferably transparent or translucent. The holding element may be a transparent slide, for example a glass slide, or a fluidic chamber. In the example shown in FIG. 1A, the holding element 15 is a Gene Frame® type fluidic chamber with a thickness e=250 μm. The thickness e of the sample 10, along the propagation axis Z, typically varies between 10 μm and 1 cm, and is preferably between 20 μm and 1 mm. The sample 10 is held on a support 10s at a distance d from an image sensor 16. The distance D between the light source 11 and the fluidic chamber 15 is preferably greater than 1 cm. It is preferably between 2 and 30 cm. Advantageously, the light source, seen by the sample, is considered to be point-like. This means that its diameter (or its diagonal) is preferably less than one tenth, better still one hundredth of the distance between the fluidic chamber 15 and the light source. In Figure 1, the light source is a light-emitting diode. It is generally associated with a diaphragm 18, or spatial filter. The aperture of the diaphragm is typically between 5 μm and 1 mm, preferably between 50 μm and 500 μm. In this example, the diaphragm is supplied by Thorlabs under the reference P150S and its diameter is 150 μm. The diaphragm can be replaced by an optical fiber, a first end of which is placed facing the light source 11 and a second end of which is placed opposite the sample 10.The device shown in Figure 1 also comprises a diffuser 17, arranged between the light source 11 and the diaphragm 18. The use of such a diffuser makes it possible to overcome constraints of centering the light source 11 relative to the opening of the diaphragm 18. The function of such a diffuser is to distribute the light beam, produced by an elementary light source 11 according to a cone of angle a. Preferably, the diffusion angle a varies between 10° and 80°. Alternatively, the light source may be a laser source, such as a laser diode. In this case, it is not recommended to associate a spatial filter or a diffuser with it, as described in document US2017317125. . Preferably, the emission spectral band AX of the incident light wave 12 has a bandwidth of less than 100 nm. Spectral bandwidth means a width at half-height of said spectral band. According to one embodiment, the light source 11 comprises several elementary light sources 11k, each being configured to emit an incident light wave 12k in a spectral band AÂk. Such a light source is shown in FIG. 1B. Preferably, the spectral bands AÀk of the different light sources 11k are different from each other. Such a configuration is described in document WO2016189257. The sample 10 is arranged between the light source 11 and an image sensor 16. The image sensor 16 is configured to form an image / 0 of the sample 10 along a detection plane Po. In the example shown, it is an image sensor comprising a pixel matrix, of the CCD or CMOS type. The pixel matrix forms the detection plane Po. The detection plane Po preferably extends perpendicular to the propagation axis Z of the incident light wave 12. The detection plane extends along a radial plane XY, defined by two axes X and Y, the radial plane being perpendicular to the propagation axis Z. The distance d between the sample 10 and the pixel matrix of the image sensor 16 is preferably between 50 pm and 2 cm, preferably between 100 pm and 2 mm. In this embodiment, we note the absence of magnification or image formation optics between the image sensor 16 and the sample 10. This does not prevent the possible presence of focusing microlenses at each pixel of the image sensor 16, the latter not having a function of magnifying the image acquired by the image sensor, their function being to optimize the efficiency of light collection by the pixels. Under the effect of the incident light wave 12, the particles present in the sample can generate a diffracted wave 13, capable of producing, at the detection plane Po, interference, in particular with a part of the incident light wave 12' transmitted by the sample. Furthermore, the sample can absorb a part of the incident light wave 12. Thus, the light wave 14, transmitted by the sample, and to which the image sensor 16 is exposed, designated by the term "exposure wave", can comprise: a component 13 resulting from the diffraction of the incident light wave 12 by each particle of the sample; a component 12' resulting from the transmission of the incident light wave 12 by the sample, a part of the latter being able to be absorbed in the sample. These components form interferences in the detection plane. Also, the image acquired by the image sensor includes interference patterns (or diffraction patterns), due to the different particles in the sample. A processor 20, for example a microprocessor, is configured to process each image / 0 acquired by the image sensor 16, according to the steps described below. In particular, the processor is a microprocessor connected to a programmable memory 22 in which is stored a sequence of instructions for carrying out the image processing and calculation operations described in this description. The processor can be coupled to a screen 24 allowing the display of images acquired by the image sensor 16 or calculated by the processor 20. Due to the absence of imaging optics, an image / 0 acquired by the image sensor 16, also called a hologram, does not provide a sufficiently accurate representation of the observed sample. The acquired image / 0 may in particular include a large number of interference patterns, and may not be easily usable for identifying the particles present in the sample. Figure 2A shows such an image, the sample being formed of polystyrene beads of diameter 5 μm bathed in water. Figure 2A was obtained by placing a sample 10 at a distance of 700 μm from an image sensor 16 of the 8-bit monochrome CMOS sensor type comprising 3884 x 2764 pixels, each pixel measuring 1.67 μm on each side. The light source was a light-emitting diode of the Créé MC-E Color type, comprising three light-emitting diodes which can be simultaneously or successively activated, each diode emitting respectively in the following spectral bands AX: 450 nm - 465 nm; 520 nm -535 nm; 620 nm - 630 nm. Only the spectral band 450 nm - 465 nm was implemented. The light source was placed at a distance of 5 cm from the sample. A holographic propagation operator h can be applied to the image / 0 acquired by the image sensor, so as to calculate a quantity representative of the exposure light wave 14. It is then possible to reconstruct a complex expression A of the light wave 14 at any point with coordinates (x,y,z) in space, and in particular in a reconstruction plane Pz located at a distance \z\ from the image sensor 16, called the reconstruction distance, this reconstruction plane being, for example, the plane Pw along which the sample extends, with: A(x,y,z) = I0(x,y,z) * h (1) * denoting the convolution product operator. The propagation operator has the function of describing the propagation of light between the image sensor 16 and a point with coordinates (x,y,z), located at a distance | z | from the sensor image. It is then possible to determine the module M(x, y, z) and / or the phase <p (x, y, z) l'onde lumineuse 14, à la distance |z|, avec : M(x,y,z) = abs [A(x,y,z)] ; <p(x,y,z) = arg [A(x,y,z)]. The operators abs and arg denote the module and the argument respectively. The propagation operator is for example the Fresnel-Helmholtz function, such that: n{x,y,z)=—e >iexp(i7r——) with i = -1. ]AZ AZ In other words, the complex expression A of the light wave 14, at any point with coordinates (x,y,z) in space, is such that: A(x,y,z) = M(x,y, In the remainder of this description, the coordinates (x,y) designate a radial position in a radial plane XY parallel to the detection plane. The coordinate z designates a coordinate along the propagation axis Z. The complex expression A is a complex quantity whose argument and modulus are respectively representative of the phase and intensity of the exposure light wave 14 detected by the image sensor 16. The convolution product of the image / 0 by the propagation operator h makes it possible to obtain a complex image Az representing a spatial distribution of the complex expression A in a reconstruction plane Pz, extending at a distance | z | from the detection plane Po. In this example, the detection plane Po has the equation z = 0. The complex image Az corresponds to a complex image of the sample in the reconstruction plane Pz. The image Az is defined at radial coordinates (x,y), such that Az(x,y) = A(x,y,z). The image Az also represents a two-dimensional spatial distribution of the complex expression of the exposure wave 14.Such a method, referred to as holographic reconstruction, makes it possible in particular to reconstruct an image of the modulus or phase of the exposure light wave 14 in the reconstruction plane. For this, images Mz and <pz représentant respectivement le module ou la phase de l'image complexe Az, avec Mz = mod (Az) et <pz = arg(Xz). . The application of a holographic reconstruction operator to an acquired image, according to expression (1), may be accompanied by the formation of noise affecting the reconstructed images. In order to limit the appearance of this noise, the application of the holographic reconstruction operator is carried out according to iterative holographic reconstruction algorithms. Such algorithms are for example described: in document WO2016189257, in particular in steps 100 to 150 described therein; or in document WO2017162985, and more precisely according to steps 110 to 160 described in the latter. The image shown in Figure 2B was obtained by applying a holographic reconstruction algorithm, as described in WO2017162985, to the image shown in Figure 2A. Figure 2B shows an image of the modulus Mz of a reconstructed complex image Az, the reconstruction distance z corresponding to a distance between a mid-plane of the sample Pio and the image sensor 16. The mid-plane of the sample Pio extends parallel to the detection plane, along the mid-thickness of the sample. Such an image allows acceptable observation of the sample. However, some particles, designated by an arrow, appear in the form of concentric rings. Such particles are located at a distance from the image sensor different from the considered reconstruction distance. They are far from the mid-plane Pio. Therefore, they do not appear clearly on the reconstructed image in the mid-plane.Thus, the images obtained from the modulus or phase of the exposure light wave 14 allow acceptable observation of the sample 10, and in particular of the particles 10p located at a distance from the image sensor 16 corresponding to the reconstruction distance. However, the particles located at a distance from the image sensor 16 greater or less than the reconstruction distance do not appear clearly on these images. The inventors have developed a method to remedy this, the main steps of which are described below, in connection with Figure 3. Step 100: Acquisition of an image / 0 of the sample 10 by the image sensor 16, this image forming a hologram. One of the advantages of the lensless configuration, shown in Figure 1A, is the wide observed field, making it possible to address a large sample volume simultaneously. The observed field depends on the size of the image sensor, being slightly smaller than the detection surface of the latter, due to the spacing between the pixels of the sensor and the sample. The observed field is generally greater than 10 mm2, and is typically between 10 mm2 and 50 mm2, which is significantly higher than with a conventional microscope. Step 110: Obtaining a stack of reconstructed images. This step involves a reconstruction of several complex images by performing holographic reconstructions at different reconstruction distances z; from the detection plane Po. We then obtain a stack of complex images Az.. Each complex image Az. of the image stack is reconstructed at a reconstruction distance Zj, such that z / =1 < Zj < Zj=j, with: z / =1 representing a minimum distance from the image sensor, along the Z axis, at which a complex image Az is reconstructed; Zj=j representing a maximum distance from the image sensor, along the Z axis, at which a complex image AZ.=J ​​is reconstructed. Each index j is a natural integer indexing a reconstruction distance, with 1 < j < J, J being the number of complex images Az. in the image stack. Preferably, the reconstruction distances Zj surround the sample 10. If the sample extends, along the propagation axis Z, between two distances zmin and zmax, relative to the image sensor 16, with zmin < zmax, the minimum reconstruction distances z;=1 and maximum Zj=j are such that Z;=1 < zmin < zmax — zj=j- The respective reconstruction distances Zj, Zj+1 of two adjacent reconstructed images Az., Az.+ [ can be separated by a distance between 5 pm and 100 pm, this distance corresponding to a reconstruction step. Each complex image Az. of the image stack can be obtained from the image / 0 acquired during step 100, according to the different reconstruction distances Zj=1 ...z} ....Zj=j. Preferably, each complex image Az. is reconstructed by applying a reconstruction algorithm limiting the reconstruction noise, as previously described. For example, an algorithm described in WO2017162985 can be used. By "from", we mean from the acquired image / 0 itself or from an image obtained by processing the acquired image / 0, for example a normalization or taking into account the square root of the image. The complex image stack can be established according to the method described in WO2017178723. At the end of step 110, a stack of complex images Az. is available, each complex image being representative of the exposure light wave 14, in a reconstruction plane located at a distance Zj from the detection plane Po. Each complex image Az. is defined according to radial coordinates (x,y), in a plane parallel to the detection plane Po. Figure 2D represents some images, called elementary images, obtained from a stack of reconstructed complex images, at different reconstruction distances, from the image in Figure 2A. The reconstruction step is 10 pm. Step 120: Calculation of a sharpness indicator for different radial positions (x, y), along the propagation axis Z. During this step, an evolution of a sharpness indicator Q(x, y) is determined along the propagation axis Z, and this for several radial positions (x, y). In other words, for each radial position (x,y), this step involves a calculation of a sharpness indicator Q(x, y), at the radial position considered, of each image Az. of the image stack. We thus obtain an evolution of the sharpness indicator Q(x,y) as a function of the reconstruction distance Zj. Each reconstructed image Az. is a complex image. The sharpness indicator Q(x, y) of each image Az. of the image stack, in a radial coordinate (x,y), can be calculated by considering an image, called elementary image Iz , obtained from the complex image. The elementary image can be: an image of the module Mz. of each complex image Az., with Iz. (x,y) = Mz (x,ÿ) = mod(Az.(x,y')) ; an image of the phase <pz. de chaque image complexe Az., avec Iz. (x,y) = (pZj(x,ÿ) = arg(AZj(x,y)) ; an image of the real part Rez. of each complex image Az., with / ?ez^.(x, y) = Æe(.4z.(x,y)), Re denoting the "real part" operator; an image of the imaginary part Imz. of each complex image Az., with Rez.(x,y) = / m(XZj.(x,y)), Im denoting the "imaginary part" operator. Generally speaking, an elementary image Iz is an image formed from scalar quantities obtained from a reconstructed complex image Az.. An elementary image allows a visualization of the complex image from which it was obtained. An elementary image can in particular be established from the modulus and / or the phase and / or the real part and / or the imaginary part of a reconstructed complex image Az.. The sharpness indicator Q(x, y) of each complex image Az. can be established from the elementary image Iz.. For example, the elementary image Iz is an image Mz. of the modulus of the complex image. The sharpness indicator can then be established by considering a gradient of the elementary image along one or more directions, and for example along the two orthogonal axes X and Y defining the radial plane XY. In this example, the sharpness indicator Cj(x,y) is obtained by: calculating a gradient of the image, along the X axis, of the module Mz. at each radial position (x,y), so as to obtain a gradient image along the X axis, denoted GxQzj (Xy)X with, in this example, Gx(IZj (x,y)) = GX{MZ. (x,y)). The gradient image can be obtained by applying a Sobel filter to the image Iz. Calculating a gradient of the image, along the Y axis, of the module Mz. at each radial position (x,y), for example by applying a Sobel filter, so as to obtain a gradient image along the Y axis, noted GY(JZ] (x>y)), with, in this example, GyQzj (Xy)) = GY(MZj (x,y)). Combining the gradient images thus calculated, so as to form a two-dimensional gradient image GXY- The two-dimensional gradient image GXY can be obtained by calculating a norm, such that: GXY(x,y) = \gx(MZj (x,y))| + \gy(MZj (x,y))| (norm of order 1) or Gxy(x,y) = ^Gx(MZj (x,y))| + ^GY(MZj (x, y))| (norm of order 2) The two-dimensional gradient image then forms, at each radial position (x,y), a sharpness indicator Q(x, y) of the complex image Az., such that Cj(x,y) = GXY(x,y). The two-dimensional gradient image thus obtained is representative of a two-dimensional gradient of the reconstructed complex image Az.. Although described in relation to an image of the modulus of the complex image, the sharpness indicator can be obtained with another component of the complex image, for example a phase image of the reconstructed complex image Az., or of its real part, or of its imaginary part. The sharpness indicator can be established, at each reconstruction distance and at each radial position, from an elementary image comprising the modulus, and / or the phase, and / or the real part, and / or the imaginary part of the reconstructed complex images, by considering the intensity of these images. The sharpness indicator can be established from the intensity of such images. Following step 120, we have a sharpness indicator Q(x, y) for each radial position (x,y), corresponding to a pixel of the complex image Az., and this for each reconstruction distance Zj. Step 130: Selecting an optimal sharpness indicator Cj(x,y) for each radial position (Xy). During this step, for each radial position (x,y), among the sharpness indicators Cj(x,y) (1 < j < / ) corresponding to the radial position, an optimal sharpness indicator Cj,(x,y) satisfying a predetermined selection criterion is selected. The optimal sharpness indicator Cj,(x,y) thus selected can be the maximum sharpness indicator, in which case Cjr(x,y) = maxC / x, y). It can also be the minimum sharpness indicator, in which case J i J Cj* (x, y) = min Cj (x, y). In the stack of elementary images illustrated in Figure 2D, radial positions in which the sharpness indicator is optimal have been framed in some elementary images. The framed radial positions correspond to particles. Step 140: Determination of an optimal reconstruction distance Zj, for each radial position (x,y) considered. The selection of the optimal sharpness indicator Cp(x, y) makes it possible to establish, for each radial position, a reconstruction distance, called optimal z / f(x,y), associated with the optimal sharpness indicator. At this optimal reconstruction distance zp(x, y), for the radial position (x,y) considered, the complex image Az., is considered to be sharp. Thus, if we consider that an image Az., at a radial position (x,y), is sharp when the sharpness indicator is maximum, we obtain, for each radial position (x, y) an optimal reconstruction distance z / f(x, y), such that: j *= argmax (Cj(x,yŸy When we consider Az., at a radial position (x,y), is sharp when the sharpness indicator is minimal, we obtain, for each radial position (x,y) an optimal reconstruction distance z / f(x, y), such that: j *= argmin [Cj(x,y)j. At the end of this step, we have a list of three-dimensional positions (x, y, z / f(x, y)), for which we consider that the observation of the sample is clear. Step 150: Formation of an observation image I of the sample. In this step, a sharp image of the sample is formed from the stack of complex images, considering the three-dimensional positions (x,y,z;*(x,y)) at which a complex image Az.t is considered sharp. The observation image I of the sample can be formed from the modulus of the complex images in the stack. In this case, the observation image of the sample is such that I(x,y) = mod(Xz. (x,y)). The observation image I of the sample can be formed from the phase of the complex images in the image stack. In this case, the observation image of the sample is such that I(x,y) = arg(Az (x,y)). The observation image I can also be obtained from the real part or the imaginary part of the complex images forming the image stack. Figure 2C corresponds to an observation image I of the sample, obtained from a stack of complex images reconstructed on the basis of the acquired image / 0 shown in Figure 2A. Some elementary images, obtained from the module of the complex images forming the stack of complex images, are shown in Figure 2D. The observation image was obtained by considering the module of the complex images forming the stack of complex images. Figure 2C corresponds to an implementation of the invention, while Figure 2B corresponds to the prior art. It can be seen that the invention allows more precise observation of the particles constituting the sample, although the latter are spatially dispersed, in particular along the propagation axis Z. Figures 4A and 4B are images of a sample taken using a microscope, respectively according to two different focusing planes. The sample contains yeasts of the Saccharomyces cerevisiae type bathed in a "YPD" culture medium known to those skilled in the art. The yeasts appearing clearly are framed. The sample is maintained in a fluidic chamber with a thickness of 250 μm. An image / 0 of the sample was acquired using a device as shown in Figure 1A. Figure 4C corresponds to an elementary image, corresponding to the module of a complex image reconstructed at a reconstruction distance corresponding to the median plane Pio of the sample. From the reconstructed complex image at the midplane of the sample, a stack of complex images was formed, with a reconstruction distance range of -200 pm to +200 pm from the midplane of the sample, with a step size of 10 pm.Figure 4D corresponds to an observation image I of the sample, obtained by implementing the invention, considering the modulus of the complex images of the image stack. We observe that: . to obtain a correct observation of the sample under the microscope, several images must be acquired, according to different focal planes; the reconstruction according to the prior art only gives a partial image of the sample, certain particles being poorly represented, in particular the particles framed in figure 4B; the observation image I according to the invention makes it possible, from a single acquisition / 0, to obtain a correct representation of the particles of the sample, and this whatever their distance from the image sensor 16 (see figure 4D); The observation image I allows, from a single image, to be able to count the particles observed, or to determine their respective geometric shapes or aspects. It is also possible to count particles according to their shapes. Figures 5A and 5B were obtained using a sample comprising PET (Polyethylene terephthalate) beads of diameter 5 pm bathed in water. An image / 0 of the sample was acquired using a device as shown in Figure 1A. Figure 5A is an elementary image obtained by considering the modulus of a complex image obtained by reconstruction from the acquired image / 0 of the sample, considering a reconstruction distance corresponding to the median plane of the sample. From the complex image reconstructed at the median plane of the sample, a stack of complex images was formed, according to a reconstruction distance range of - 400 pm to + 200 pm from the median plane of the sample, with a reconstruction step of 10 pm between two adjacent reconstructed images.Figure 5B corresponds to an observation image I of the sample, obtained by implementing the invention, considering the modulus of the complex images of the image stack. It is observed that in Figure 5A, certain particles do not appear clearly, because they are far from the considered reconstruction distance. In Figure 5B, all the particles appear correctly, so as to be able to count them or estimate their shape. According to one embodiment, the method comprises an optional step 145, according to which a selection is made, for each radial coordinate (x, y) considered, of the optimal distance zM(x, y) obtained following steps 130 and 140. According to this selection, when, for a radial position (x, y), the optimal distance is located outside a previously determined validity range Az, it is invalidated. The validity range Az corresponds for example to a range of distances [zTO,„, zmnr] delimiting the sample. Thus, if z,*(x, y) < zm,„ or Zjr(x,y) > zmax, the optimal distance z^Çx.y) is invalidated. The sharpness indicator Cj(x,y) corresponding to such a distance is then invalidated. For this radial position, steps 130 and 140 can then be repeated. Such an embodiment makes it possible to prevent elements located outside the sample, for example dust placed on the holding element 15, or on the image sensor 16, from being considered as part of the sample 10. According to such an embodiment, step 150 is optional. In this case, the method makes it possible to obtain an optimal distance at each radial position considered, without necessarily resulting in an observation image of the sample. According to one variant, an image-forming optic is arranged between the sample and the image sensor. According to one variant, illustrated in FIG. 6, the device comprises an optical system 19, defining an object plane PObj and an image plane Pim. The image sensor 16 is then arranged in a so-called defocused configuration, according to which the sample extends along a plane offset from the object plane, and / or the image sensor extends along a plane offset from the image plane. By defocused configuration, is meant a configuration comprising an offset of the sample and / or the image sensor relative to a focusing configuration, according to which the detection plane Po is conjugated with a plane Pio along which the sample extends. The offset ô is preferably less than 500 pm, or even 200 pm. It is preferably greater than 10 pm or 20 pm.In the same way as in the lensless configuration, such a configuration makes it possible to obtain an image in which each microorganism appears in the form of a diffraction pattern, interference occurring between the light wave emitted by the light source and propagating to the image sensor and a diffraction wave generated by each particle of the sample. In the example shown in Figure 6, the object plane PObj is merged with a plane Pio along which the sample extends and the image plane Pim is offset relative to the detection plane Po. The method described in connection with steps 100 to 150 is applicable to images acquired according to such a configuration. However, a lensless imaging configuration is preferred, due to the larger field of observation that it provides. The invention can be implemented, in a non-limiting manner, in the field of diagnosis, biology, or in environmental control, or even in the field of agri-food or industrial process control.

Claims

CLAIMS 1. Method for observing a sample (10), comprising the following steps: a) illuminating the sample using a light source (11), the light source emitting an incident light wave (12) propagating towards the sample (10) along a propagation axis (Z); b) acquisition, using an image sensor (16), of an image ( / 0) of the sample (10), formed in a detection plane (Po), the sample being arranged between the light source (11) and the image sensor (16), each image being representative of an exposure light wave (14), to which the image sensor (16) is exposed under the effect of the illumination, the image comprising pixels, each pixel corresponding to a radial position (x, y) defined in a plane parallel to the detection plane; c) forming a stack of complex images, called reconstructed images, from the image acquired ( / ) during step b), each reconstructed image being obtained by applying a digital propagation operator (h), according to a reconstruction distance (Zj), along the propagation axis (Z), the stack of images comprising as many reconstructed images (Az.) as there are different reconstruction distances, each reconstructed image (Az.) being representative of an exposure light wave (14) to which the image sensor (16) is exposed; d) from each image (Az.) of the image stack, calculation of a sharpness indicator (C^x, y)) for different radial positions (x,y), each sharpness indicator being associated with a radial position (x,y) and a reconstruction distance (zp; (e) taking into account a selection criterion; f) for each radial position (x,y), and depending on the selection criterion, selection of a sharpness indicator (C / t(x,y)), from among the different sharpness indicators (Cj(x,y)) defined, at the radial position (x,y), from the different reconstructed images (AZj), the selected sharpness indicator (C / t(x,y)) being associated with an optimal reconstructed image ( / 1^) for the radial position considered; g) forming an observation image ( / ) of the sample, each pixel of which is associated with a radial position (x,y), the pixel value of the observation image being determined as a function of the value, at the radial position of the pixel, of the optimal reconstructed image (A^.*) associated with it; the method being characterized in that during step d): each sharpness indicator (C;(x,y)) is calculated from an elementary image (Iz.), the elementary image being established from the module and / or the argument and / or the real part and / or the imaginary part of a reconstructed image (Az.); each sharpness indicator is established by considering a gradient of the elementary image (Iz.) according to one or more gradient directions (X,Y).

2. Method according to claim 1, comprising, following step f), a determination, for each radial position (x,y) considered, of an optimal distance (z / t(x, y)), corresponding to the reconstruction distance of the optimal reconstructed image (Az.t) associated with the radial position.

3. Method according to any one of the preceding claims, in which step d) comprises, for each reconstructed image (Az.) of the image stack: a calculation of a gradient image (GX(IZ. (x, y)), GY(IZj (x, y))), representative of a gradient of the reconstructed image (Az.) according to at least one gradient direction (X,Y); a calculation of a norm of the gradient image; such that the sharpness indicator (Q / x, y)), at each radial position (x,y), is obtained from the norm of the gradient image at the radial position.

4. Method according to any one of the preceding claims, in which step f) comprises taking into account a validity range (Az), and rejecting a selected sharpness indicator (C / t(x, y)) if the reconstruction distance (z / t(x, y)) associated with it is not included in the validity range (Az).

5. A method according to any preceding claim, wherein no imaging optics are disposed between the sample and the image sensor.

6. Method according to any one of claims 1 to 4, in which an optical system (19) is arranged between the sample and the image sensor, the optical system combining an image plane (Pim) with an object plane (Pobj), the detection plane being offset relative to the image plane, and / or the sample being offset from the object plane, such that the image acquired by the image sensor is a defocused image of the sample (10).

7. A method according to any preceding claim, wherein the sample is held in or on a holding member (15), the sample being stationary relative to the holding member.

8. Method according to any one of the preceding claims, in which the sample comprises particles (10p), the method comprising a step h) of characterizing the particles from the observation image ( / ) of the sample, the characterization comprising: a particle count; and / or a determination of a size and / or shape of the particles; and / or a counting of particles based on their size and / or shape; and / or a determination of a three-dimensional position of the particles.

9. Device for identifying a sample, comprising: - a light source (11) configured to emit an incident light wave (12) propagating towards the sample (10); - an image sensor (16); - a support (10s), configured to hold the sample (10) between the light source (11) and the image sensor (16); a processor (20), configured to receive an image ( / 0) of the sample acquired by the image sensor (16) and to implement steps c) to g) of a method according to any one of claims 1 to 8.

10. Device according to claim 9, wherein no imaging optics are arranged between the sample (10) and the image sensor (16).

11. Device according to claim 9, comprising an optical system (19) defining a focusing configuration, according to which the image sensor (16) is conjugated with a plane passing through the sample, the device being such that the image sensor (16) or the sample (10) are offset relative to the focusing configuration, such that during step b), the image sensor (16) acquires a defocused image of the sample (10).

Images