Beam profile analysis in conjunction with a tof sensor

Abstract

A method for determining whether an object corresponds to a living organism, comprising: i) (132) illuminating the object with a light beam pattern generated by at least one emitter device (116) of a time-of-flight system (114), wherein the emitter device (116) comprises at least one array of light emitters (118); ii) (134) generating at least one pattern image (122) of the object while the object is illuminated by the light beam pattern by using at least one detector (120) of the time-of-flight system (114); iii) (136) extracting living data from the pattern image (122) by using beam profile analysis, and determining whether the object corresponds to a living organism based on the living data.

Description

Technical Field

[0001] The present invention relates to a method for determining whether an object corresponds to a living organism, a system for determining whether an object corresponds to a living organism, a computer-implemented method for authenticating a user of an authenticating device, a system for authenticating a user, a computer program, a computer-readable storage medium, a non-transient computer-readable medium, and several uses.

[0002] The apparatus, method, and uses according to the invention can be specifically applied, for example, in various fields of daily life, security technology, gaming, transportation technology, production technology, photography (e.g., digital or video photography for artistic, documentation, or technical purposes), safety technology, information technology, agriculture, crop protection, maintenance, cosmetics, and medical technology, or in science. However, other applications are also possible. Background Technology

[0003] 3D detectors, including time-of-flight (TOF) sensors, are well-known and can be used in mobile devices such as smartphones and tablets for distance estimation. Authentication systems for mobile devices can test for deception, such as determining whether an object corresponds to a living organism, and require specific and additional sensors for these tests, such as material detection. There is still a need to save on hardware space and cost, and to reduce the number of components used in authentication systems for mobile devices.

[0004] WO 2023 / 287351 describes a three-dimensional (3D) facial recognition system. The system includes a structured light sensor or stereo camera sensor, a time-of-flight sensor, and a processor. The processor is configured to run an algorithm for facial recognition on data from the structured light sensor or stereo camera sensor, wherein the processor is configured to optimize the algorithm for facial recognition using distance data from the time-of-flight sensor. A 3D facial recognition method and associated computer program products are also disclosed.

[0005] US 11,099,009 describes an imaging apparatus including a circuit system configured to perform depth analysis on a scene via time-of-flight imaging and motion analysis on the scene via structured light imaging, wherein the same sensor data is used for both depth analysis and motion analysis.

[0006] US10,613,228 describes a depth camera that uses structured light and modulated light to generate a depth image. Modulated and structured light allow for the calculation of time-of-flight depth for each cell (e.g., a single point) of a reflected structured light image captured by the depth camera's image sensor. Once each cell of the reflected structured light image is identified, a structured light triangulation algorithm can be used to calculate the depth of each cell of the reflected structured light image.

[0007] Li Larry provides an overview of Texas Instruments' time-of-flight camera in "Time-of-Flight Camera – An Introduction", May 31, 2014, XP093183136, www.ti.com / lit / wp / sloa190b / sloa190b.pdf?ts=1720418472608.

[0008] US 11,468,712 B2 describes a liveness detection device and its authentication method. The liveness detection device includes a light source unit, an image sensor unit, and a data processing module. The light source unit includes a substrate having a first tilted surface, thereby emitting light reflected from the first tilted surface. An application triggers an authentication process, which is instructed to the user. The light source unit begins illumination with a specific pattern for a specific period, generating an image signal. A liveness detection signal is generated by sequentially calculating interference patterns based on more than one image signal to determine liveness. When a liveness threshold is met, feature recognition data is generated by sequentially calculating interference patterns based on more than one image signal for matching. The features are then compared with previously registered data to lock or unlock the liveness detection device and / or the system coupled thereto.

[0009] US 2021 / 181305 A1 describes a liveness testing method and a liveness testing apparatus. The liveness testing method includes: determining the presence of a subject using a radar sensor; performing a first liveness test on the subject based on radar data obtained from the radar sensor in response to the presence of the subject; acquiring image data of the subject using an image sensor in response to the result of the first liveness test satisfying a first condition; and performing a second liveness test on the subject based on the image data. The problem to be solved

[0010] Therefore, the object of the present invention is to provide devices and methods that address the aforementioned technical challenges posed by known devices and methods. Specifically, the object of the present invention is to provide methods and devices for authenticating users (e.g., their mobile devices), thereby allowing for a reduction in the number of components to save hardware space and cost. Summary of the Invention

[0011] This problem is solved by methods for determining whether an object corresponds to a living organism, systems for determining whether an object corresponds to a living organism, computer-implemented methods for authenticating a user of a device, systems for authenticating a user, computer programs, computer-readable storage media, and non-transient computer-readable media, all having the features of the independent claims. Advantageous embodiments that can be implemented independently or in any arbitrary combination are set forth in the dependent claims and throughout the specification.

[0012] In the first aspect, a method for determining whether an object corresponds to a living organism is disclosed.

[0013] As used herein, the term "object" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or customized meaning. The term can specifically refer to, but is not limited to, any target selected from living and non-living objects. An object can be or may include one or more organisms and / or one or more parts thereof, such as one or more body parts of a human (e.g., a user). An object can be a non-living object, such as a silicone mask or a printed image of a human. As used herein, the term "living organism" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or customized meaning. The term can specifically refer to, but is not limited to, any living organism, particularly a living human. As used herein, the term "living human" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or customized meaning. The term can specifically refer to, but is not limited to, an individual of the Homo sapiens species, wherein the individual is currently alive.

[0014] Determining whether a user corresponds to a living organism can include distinguishing between non-living organisms, such as distinguishing between skin materials and non-skin materials (like silicone or paper).

[0015] The method steps can be executed in a given order. However, a different order is also possible. Furthermore, two or more of these method steps can be executed simultaneously. Thus, these method steps can at least partially overlap in time. Furthermore, these method steps can be executed once or repeatedly. Furthermore, one or more, or even all, of these method steps can be executed once or repeatedly. The method may include additional method steps not listed herein.

[0016] The method includes:

[0017] i) Illuminate the object with a beam pattern generated by at least one emitting device of the time-of-flight system, wherein the emitting device includes at least one array of light emitters;

[0018] ii) By using at least one detector of the time-of-flight system, at least one pattern image of the object is generated while the object is illuminated by the beam pattern;

[0019] iii) Extract live data from the pattern image using beam profile analysis, and determine whether the object corresponds to a living organism based on the live data.

[0020] As used herein, the term "light" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or customary meaning. Specifically, the term may refer to, but is not limited to, electromagnetic radiation in one or more of the infrared, visible, and ultraviolet spectral ranges. In this document, the term "ultraviolet spectral range" generally refers to electromagnetic radiation with wavelengths from 1 nm to 380 nm, preferably from 100 nm to 380 nm. Further, in part according to the standard ISO-21348, the effective version of this document as of the date of this document, the term "visible spectral range" generally refers to the spectral range from 380 nm to 760 nm. The term "infrared spectral range" (IR) generally refers to electromagnetic radiation from 760 nm to 1000 µm, wherein the range from 760 nm to 1.5 µm is generally referred to as the "near-infrared spectral range" (NIR), the range from 1.5 µm to 15 µm is referred to as the "mid-infrared spectral range" (MidIR), and the range from 15 µm to 1000 µm is referred to as the "far-infrared spectral range" (FIR). Preferably, the light used for the typical purposes of this invention is light in the infrared (IR) spectral range, more preferably light in the near-infrared (NIR) and / or mid-infrared spectral range (MidIR), especially light with wavelengths from 1 µm to 5 µm, preferably from 1 µm to 3 µm. For example, the light beam may have wavelengths from 760 nm to 1.5 µm, preferably 940 nm, 1140 nm, or 1320 to 1380 nm. Other wavelengths are also possible.

[0021] As used herein, the term "ray" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom-defined meaning. Specifically, the term may refer to, but is not limited to, a line perpendicular to the wavefront of light and pointing in the direction of energy flow. As used herein, the term "beam" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom-defined meaning. Specifically, the term may refer to, but is not limited to, a collection of rays. In the following text, the terms "ray" and "beam" will be used as synonyms. As used herein, the term "beam" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom-defined meaning. Specifically, the term may refer to, but is not limited to, a quantity of light, specifically a quantity of light traveling substantially in the same direction, including the possibility that the beam has an extension angle or widening angle.

[0022] As used herein, the term "irradiation" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, the process of exposing at least one element to light. Irradiation may include projecting a pattern onto a scene including an object (e.g., a face).

[0023] As used herein, the term "pattern" (also referred to as a light pattern) is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, at least one arbitrary pattern comprising multiple light beams (at least two beams, preferably at least two beams). The pattern may be projected onto an object. Projecting a light beam of the light pattern onto a surface can produce a light spot. The light beam may illuminate at least a portion of the surface. The light spot may refer to a continuous region of coherent electromagnetic radiation on at least a portion of the surface. The light spot may refer to a coherent electromagnetic radiation spot of any shape. The light spot may be the result of projecting a light beam associated with the light pattern.

[0024] The light spots can extend at least partially in space. The emitted light pattern can illuminate the surface by a light pattern comprising multiple light spots. The light spots can at least partially overlap. For example, the number of light spots can be equal to the number of beams associated with the emitted light pattern. The intensities associated with the light spots can be substantially similar. Substantially similar can mean that the intensity values ​​associated with the light spots can differ by less than 50%, preferably less than 30%, more preferably less than 20%. Using patterned light may be advantageous because it allows for the avoidance of photosensitive areas such as the eyes. The pattern can include at least one dot pattern.

[0025] As used herein, the term "system" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, any set of interacting or interdependent components forming a whole. Specifically, these components may interact with each other to achieve at least one common function. At least two components may be processed independently, or may be coupled or connected.

[0026] As used herein, the term "Time-of-Flight (TOF) system" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, any system configured to measure the distance between a detector and an object based on time-of-flight. A TOF system may include at least one transmitting device and at least one detector, particularly at least one phase detector imager.

[0027] As used herein, the term "emitting device" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, any device configured to generate or provide light as defined above. An emitting device may be and / or may include at least one illumination source. An emitting device includes at least one array of light emitters.

[0028] The array can be two-dimensional or one-dimensional. The array can include multiple light emitters arranged in a matrix. The light emitters of the array can be arranged as a one-dimensional (e.g., row) array or a two-dimensional array, particularly as a matrix with m rows and n columns, where m and n are independently positive integers. As used herein, the term "matrix" is a general term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term can specifically refer to, but is not limited to, the arrangement of multiple elements in a predetermined geometric order. Specifically, a matrix can be or can include a rectangular matrix having one or more rows and one or more columns. Specifically, the rows and columns can be arranged in a rectangular manner. However, other arrangements are also possible, such as non-rectangular arrangements. The light emitters can be arranged such that the pattern is a hexagonal pattern. Therefore, for example, the matrix can be a hexagonal matrix.

[0029] As used herein, the term "light emitter" (also simply "emitter") is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, at least one arbitrary device configured to provide at least one light beam.

[0030] The transmitters can be configured to emit light pulses. Each light transmitter can be configured to emit beams one after another (e.g., within time intervals), for example, controlled by at least one control unit of the TOF system.

[0031] The control unit can be configured to control the transmitter and detector of the Time-of-Flight (TOF) system. Control may include providing control signals, particularly high-speed signals, to the transmitter and detector. Control may include synchronizing the transmitter and detector. The control unit may include driver electronics configured to perform specified operations.

[0032] The emitter may include one or more of at least one laser source, at least one light-emitting diode, or at least one laser diode. For example, the laser source may be at least one vertical-cavity surface-emitting laser (VCSEL). As used herein, the term "vertical-cavity surface-emitting laser" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a semiconductor laser diode configured to emit a laser beam perpendicularly to its top surface. Examples of VCSELs can be found, for example, at en.wikipedia.org / wiki / Verticalcavity_surface-emitting_laser.

[0033] For example, the transmitting device may be or may include at least one radio frequency (RF) modulated light source. In this case, the optical transmitter may be and / or may include at least one light-emitting diode or at least one laser diode. The transmitting device may be configured to modulate light generated by the optical transmitter using an RF carrier. The transmitting device may be configured to modulate light having an RF carrier frequency from 100 kHz to 300 MHz, preferably from 80 MHz to 200 MHz. For example, the transmitting device may be configured to modulate light at a high speed of up to 100 MHz.

[0034] For example, a TOF system may include at least one direct TOF imager. In this case, the light emitter may be and / or may include at least one laser source, such as at least one infrared laser source. This allows the illumination to be imperceptible. A single pulse per frame can be used, ranging from 10 to 100 Hz, preferably from 15 to 80 Hz, and more preferably from 20 to 60 Hz. For example, a single pulse per frame, such as 30 Hz, can be used.

[0035] The emitting device may include at least one imaging and / or collimating optics, which includes at least one optical element selected from the group consisting of: at least one refractive lens; at least one metasurface lens; and at least one diffractive optical element (DOE). The optical elements are configured to collimate and / or reproduce light. As used herein, the term "collimation" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, aligning the direction of motion of an incident (e.g., diverging) beam as parallel rays. As used herein, the term "reproducing" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, increasing (e.g., replicating) the number of light spots (e.g., light spots generated by a light emitter). The optical elements may include at least one diffractive optical element (DOE) and / or at least one metasurface element. The DOE and / or metasurface element may be configured to generate multiple beams from a single incident beam. For example, the emitter can generate up to 2000 light spots, and optical elements including multiple metasurface elements can be used to replicate the number of light spots. Additional arrangements (particularly including different numbers of projection emitters and / or at least one different optical element configured to increase the number of light spots) are possible. Other multiplication factors are also possible.

[0036] Step i) includes generating at least one patterned image of the object while the object is illuminated by the beam pattern using at least one detector of the time-of-flight system.

[0037] The light beam illuminating the object can be reflected by the object and imaged by the detector of the TOF system. As used herein, the term "detector of a time-of-flight system" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term can specifically refer to, but is not limited to, any device configured to detect incident light, particularly for TOF analysis. The detector can be and / or may include at least one image sensor. The detector includes at least one pixelated imaging element comprising a plurality of optical sensors. The optical sensors can be arranged in a two-dimensional matrix. For example, the detector may include at least one CMOS sensor or at least one CCD chip.

[0038] This invention proposes a combination of two methods using only a Time-of-Flight (TOF) detector, specifically without using any additional or supplementary photodetectors or light sensors. The detector in the TOF system is used for both TOF detection and light detection to generate a patterned image. Therefore, the TOF detector can have two functions: it can be configured to measure the time it takes for light to travel from the emitting device to the object and from the object to the detector, and it is used to detect light to generate a patterned image.

[0039] Each pixel of the detector can be configured to measure the time it takes for light to travel from the transmitter to the object and back to the detector. For example, in the case of using an RF modulated light source, the detector of a TOF system can be configured to measure the phase shift of the carrier wave. For example, in the case of using the direct time-of-flight measurement principle, the detector can be configured to determine the time it takes for a single pulse to leave the transmitter and reflect back to the detector. In particular, when operating in a so-called triggered mode, the detector can generate a 3D image that includes both spatial and temporal data. The detector (e.g., each pixel) can be configured to measure the intensity of the incident light, particularly the amplitude of the modulated light on the detector.

[0040] As used herein, the term "image" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, data recorded using a detector, such as multiple electronic readings from a CMOS or CCD chip. An image may include raw image data or may be a preprocessed image. For example, preprocessing may include applying at least one filter and / or at least one background correction and / or at least one background subtraction to the raw image data.

[0041] The detector's optical sensor can be configured to measure the intensity of light from an object under beam pattern illumination. Each pixel of the detector can be configured to measure the intensity of the incident light, particularly the intensity of the beam reflected from the object under beam pattern illumination. The detector is configured to generate a pattern image. As used herein, the term "pattern image" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term can refer to, but is not limited to, an image generated by the detector of a time-of-flight system while an object is illuminated by a beam pattern. A pattern image can include an image showing at least a portion of the object, particularly the user's face, when the user is illuminated by the pattern, particularly within a corresponding region of interest included in the image. Thus, the detector can generate an intensity image of the object projected by the pattern. The pattern image can be generated by imaging and / or recording the light reflected from the object illuminated by the light pattern.

[0042] Therefore, this invention proposes using a TOF system to generate images, particularly 2D images, such as grayscale images, for evaluation to extract live data through beam profile analysis.

[0043] In step iii), the image can be processed, for example, by using beam profile analysis, to obtain material information about the object.

[0044] Step iii) includes extracting live data from the pattern image using beam profiling analysis, and determining whether the object corresponds to a living organism based on the live data.

[0045] For example, a pattern image can be reduced to a predefined size by applying one or more image processing techniques. Reduction may include selecting at least one region of interest and segmenting the pattern image into regions of the predefined size. These predefined-sized regions of the pattern image may be associated with an object. Portions of the image outside these predefined-sized regions may be associated with the background and / or may be unrelated to the object. Parts of the pattern image that are useful for subsequent analysis can be selected.

[0046] Determining whether an object corresponds to a living organism based on a pattern image can include extracting liveness data from the pattern image. In particular, this can be used to determine whether the object is a human.

[0047] As used herein, the term "living data" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, data that provides indications that an object corresponds to a living organism. In particular, extracting living data includes extracting material data and / or extracting blood perfusion data and / or extracting at least one measure of surface roughness.

[0048] For example, determining whether an object corresponds to a living organism based on at least one pattern image may include extracting material data from the pattern image. Specifically, this can be used to determine that the user is human. The material data may include information about the type of material of the user detected in the pattern image. Extracting material data from the pattern image may or may include generating material type and / or data derived from the material type. The material data may include information about the material type of the object detected in the pattern image.

[0049] Extracting material data from patterned images can include beam profile analysis of light spots. For information on beam profile analysis, see WO 2018 / 091649 A1, WO 2018 / 091638 A1 and WO 2018 / 091640 A1, WO 2020 / 187719 A1, WO 2023 / 156469 A1, and WO 2023 / 156315 A1, the entire contents of which are incorporated herein by reference. Beam profile analysis can allow for reliable classification of objects based on several light spots. Each light spot in a patterned image can include a beam profile. As used herein, the term "beam profile" can generally refer to at least one intensity distribution of a light spot on an optical sensor as a function of pixels. Beam profiles can be selected from the group consisting of: trapezoidal beam profiles; triangular beam profiles; conical beam profiles; and linear combinations of Gaussian beam profiles.

[0050] Determining whether an object corresponds to a living organism based on at least one pattern image may include determining whether extracted material data corresponds to desired material data. Determining whether material data corresponds to desired material data may include comparing the material data with the desired material data. In an example, skin may be compared as desired material data with a non-skin material or silicon as material data, and the result may be rejection because silicon or non-skin materials may differ from skin. Comparing the material data with the desired material data may include determining the similarity between the extracted material data and the desired material data. The desired material data may refer to predetermined material data. In an example, the desired material data may be skin. It may be determined whether the material data corresponds to the desired material data. In an example, the material data may be a non-skin material or silicon.

[0051] To extract material data, a complete pattern image can be used. Alternatively, a partial image can be used. Material data extraction may include generating one or more partial images from a pattern image. For example, different regions of the pattern image can be selected as partial images. The partial images can be different from each other. In particular, the partial images can be non-overlapping. Using partial images allows for the collection of material data from different regions of an object (e.g., under different lighting conditions) and / or comparison of the extracted material data and / or generation of material maps and / or reduction of uncertainty in the obtained material data.

[0052] In an embodiment, extracting material data from a pattern image may include generating material type and / or data derived from the material type. Preferably, the extraction of material data may be based on the pattern image.

[0053] Material data can be extracted using at least one model. Extracting material data may include providing a patterned image to at least one model and / or receiving material data from the model. Providing an image to the model may include, and subsequently may be, receiving the image at the model's input layer or via a model loss function.

[0054] The model can be a data-driven model. Data-driven models can include convolutional neural networks and / or encoder-decoder structures, such as autoencoders. Other examples for generating representations include FFT, wavelets, deep learning (such as CNNs), energy models, normalized flow, GANs, visual transformers or transformers for natural language processing, and autoregressive image modeling. Supervised or unsupervised schemes can be applied to generating representations and also to generating embeddings in ML languages, such as cosine or Euclidean metrics.

[0055] The data-driven model can be parameterized based on a training dataset comprising at least one image and material data, preferably at least one pattern image and material data. Specifically, the training dataset includes multiple historical pattern images and corresponding material data. In another embodiment, extracting material data may include providing images to the model and / or receiving material data from the model. In another embodiment, the data-driven model can be trained based on a training dataset comprising at least one image and material data. In another embodiment, the data-driven model can be parameterized based on a training dataset comprising at least one image and material data. The data-driven model can be parameterized based on the training dataset to receive images and provide material data as output based on the received images. The training dataset may include at least one image and material data (preferably material data associated with at least one image).

[0056] Extracting material data may include generating a representation associated with a patterned image of an object, such as at least one tensor representing the patterned image or a dimensionality-reduced representation of the patterned image. The image may include a representation of the image. The representation may be a low-dimensional representation of the image, such as a tensor. The representation may include at least a portion of the data or information associated with the image. The image representation may include feature vectors. In embodiments, determining the representation, particularly the low-dimensional representation, may be based on principal component analysis (PCA) mapping or radial basis function (RBF) mapping. Determining the representation may also be referred to as generating the representation. Generating a representation based on a PCA mapping may include clustering based on features in the patterned image and / or a portion of the image. Alternatively or additionally, the generating representation may be based on a neural network architecture suitable for dimensionality reduction. A neural network architecture suitable for dimensionality reduction may include an encoder and / or a decoder. In an example, the neural network architecture may be an autoencoder. In an example, the neural network architecture may include a convolutional neural network (CNN). A CNN may include at least one convolutional layer and / or at least one pooling layer. A CNN may reduce the dimensionality of a portion of the image and / or the image by applying convolutions (e.g., based on convolutional layers) and / or by pooling. Applying convolution can be suitable for selecting features that are relevant to material information in parts of an image.

[0057] In embodiments, the model may be adapted to determine the output based on the input. Specifically, the model may be adapted to determine material data based on an image as input. The model may be a deterministic model, a data-driven model, or a hybrid model. Preferably, the deterministic model reflects the physical phenomenon in a mathematical form, for example, including a first-principles model. The deterministic model may include a set of equations describing the interaction between the material and patterned electromagnetic radiation, thereby producing measures of condition, vital signs, etc. The data-driven model may be a classification model. The hybrid model may be a classification model including at least one machine learning architecture and model parameters with deterministic or statistical adjustments. Statistical or deterministic adjustments may be introduced to improve the quality of the results because these adjustments provide a systematic relationship between empiricism and theory. In embodiments, the data-driven model may be a classification model. The classification model may include at least one machine learning architecture and model parameters. For example, the machine learning architecture may be or may include one or more of the following: linear regression, logistic regression, random forest, piecewise linear, nonlinear classifier, support vector machine, Naive Bayes classification, nearest neighbor, neural network, convolutional neural network, generative adversarial network, support vector machine, or gradient boosting algorithm, etc. In the case of neural networks, the model can be a multi-scale neural network or a recurrent neural network (RNN), such as, but not limited to, a gated recurrent unit (GRU) recurrent neural network or a long short-term memory (LSTM) recurrent neural network. The data-driven model can be parameterized based on a training dataset. The data-driven model can be trained based on the training dataset. Training the model can include parameterizing the model. The term "training" can also mean learning. Specifically, the term can refer to, but is not limited to, the process of building a classification model, and particularly determining and / or updating the parameters of a classification model. Updating the parameters of a classification model can also be referred to as retraining. Training as discussed herein can include retraining. In embodiments, the training dataset can include at least one image and material information.

[0058] In embodiments, extracting material data from an image using a data-driven model may include providing the image to the data-driven model. Alternatively or additionally, extracting material data from an image using a data-driven model may include generating an embedding associated with the image based on the data-driven model. The embedding may refer to a low-dimensional representation associated with the image, such as a feature vector. The feature vector may be adapted to suppress background while preserving a material signature indicating the material data. In this context, the background may refer to information independent of the material signature and / or the material data. Further, the background may refer to information related to biometric features, such as facial features. Based on the embedding associated with the image, the material data can be determined using the data-driven model. Alternatively or additionally, extracting material data from an image by providing the image to the data-driven model may include transforming the image into material data, particularly material feature vectors indicating the material data. Therefore, the material data may further include material feature vectors and / or the material feature vectors may be used to determine the material data.

[0059] Determining that an object corresponds to a living organism may involve comparing the extracted material data with expected material data. If the material data matches the expected material data at least within tolerances, the object is determined to correspond to a living organism. Otherwise, if the material data does not match the expected material data, the object is determined to correspond to a non-living organism.

[0060] Determining whether extracted material data corresponds to expected material data may include determining the similarity between the extracted material data and the expected material data. Determining the similarity between the extracted material data and the expected material data may include comparing the extracted material data with the expected material data. Expected material data may refer to predetermined material data. In an example, the expected material data may be skin. It may be determined whether material data corresponds to the expected material data. In an example, the material data may be a non-skin material or silicon. Determining whether material data corresponds to the expected material data may include comparing the material data with the expected material data. The comparison of material data with the expected material data may result in allowing and / or denying a user and / or object from performing at least one operation requiring authentication. In an example, skin may be compared as the expected material data with a non-skin material or silicon as the material data, and the result may be denial because silicon or non-skin materials may differ from skin. If the material data matches the material data of skin, the object is determined to correspond to a human. Otherwise, if the material data does not match the material data of skin, the object is determined to correspond to a non-living organism.

[0061] In an embodiment, step iii) may include generating at least one feature vector from the material data and matching the material feature vector with at least one associated reference template vector of the material.

[0062] Determining whether an object corresponds to a living organism based on at least one pattern image may include determining at least one blood perfusion measurement. In particular, this can be used to determine that the human being is alive.

[0063] As used herein, the term "blood perfusion measure" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, the blood flow through a given volume or mass of tissue. Typically, blood perfusion measure may be given in ml / ml / s or ml / 100 g / min. Blood perfusion measure can represent localized blood flow through at least one capillary network and one or more extracellular spaces in a body tissue.

[0064] Determining the at least one blood perfusion measure may include determining at least one speckle contrast of the patterned image. Alternatively or additionally, determining the at least one blood perfusion measure may include determining the blood perfusion measure based on the determined at least one speckle contrast. As used herein, the term "speckle contrast" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, the degree of variation in a speckle pattern generated by coherent light. The speckle pattern can be generated by an emitting device, particularly on an object. The speckle contrast can be a measure of the average contrast of the intensity distribution within a region of the speckle pattern. In particular, the speckle contrast K over a region of the speckle pattern can be expressed as the standard deviation σ versus the average speckle intensity. The ratio, that is,

[0065]

[0066] Speckle contrast can include speckle contrast values. These values ​​can range from 0 to 1. Blood perfusion can be determined based on speckle contrast. Blood perfusion can depend on the determined speckle contrast. If the speckle contrast changes, the blood perfusion obtained from it will also change accordingly. Blood perfusion can be a single number or value that represents the likelihood that the object is alive. To monitor changes in speckle contrast, multiple pattern images generated at different time points can be used.

[0067] To determine speckle contrast, a full-pattern image can be used. Alternatively, a portion of a pattern image can be used. Preferably, a portion of the pattern image represents a smaller area of ​​the pattern image than the full-pattern image.

[0068] In this embodiment, a data-driven model can be used to determine blood perfusion measurements. The data-driven model is parameterized and / or trained based on a training dataset. The training dataset may include pattern images and blood perfusion measurements. Specifically, the training dataset includes multiple historical pattern images and corresponding blood perfusion measurements. The data-driven model can be parameterized and / or trained based on the training dataset to output blood perfusion measurements based on received pattern images. For embodiments of the data-driven model, refer to the description of the data-driven model in the extraction of material data.

[0069] In an embodiment, determining whether an object corresponds to a living organism based on blood perfusion data may include determining whether a blood perfusion measurement corresponds to a human blood perfusion measurement. Determining whether a blood perfusion measurement corresponds to a human may include comparing the blood perfusion measurement to at least one predefined or predetermined range of blood perfusion measurement values, for example, stored in at least one database. If the extracted blood perfusion measurement falls within the redefined or predetermined range of blood perfusion measurement values, at least within tolerance, the object is determined to correspond to a living organism; otherwise, it does not.

[0070] As described above, extracting live data may include extracting at least one surface roughness measure. The surface roughness measure may be extracted from at least one speckle image of the object while it is being irradiated. For example, the object may be irradiated with coherent electromagnetic radiation associated with wavelengths between 850 nm and 1400 nm. As used herein, the term "coherent" electromagnetic radiation is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, light patterns and / or multiple beams of light having at least substantially a fixed phase relationship between electric field values ​​at different locations and / or different times. In particular, coherent electromagnetic radiation may refer to electromagnetic radiation capable of exhibiting interference effects. The term "coherent" may also include partial coherence, i.e., imperfect correlation between phase values. The electromagnetic radiation may be fully coherent, wherein a deviation of approximately ±10% in the phase relationship is possible. This coherent electromagnetic radiation is associated with wavelengths between 850 nm and 1400 nm. Preferably, the coherent electromagnetic radiation may be in the infrared range. Preferably, the coherent electromagnetic radiation can be associated with wavelengths between 880 nm and 1300 nm. Specifically, the coherent electromagnetic radiation is associated with wavelengths between 900 nm and 1000 nm, and / or with wavelengths between 1100 nm and 1200 nm. This can be advantageous because sunlight has a band gap in these ranges. Therefore, it is easier to distinguish the coherent electromagnetic radiation with the aforementioned wavelengths that is used to generate the speckle image from the incident sunlight. Therefore, using coherent electromagnetic radiation within the range specified above makes it possible to measure surface roughness even in the presence of sunlight (e.g., under natural conditions). Therefore, surface roughness measurements can be easily and location-independently used. Overall, an improved signal-to-noise ratio can be achieved, and the accuracy of surface roughness assessment can be improved.

[0071] A speckle image can include one or more speckles. Projecting patterned coherent electromagnetic radiation onto a regular surface can make the speckle independent of the speckle. Projecting patterned coherent electromagnetic radiation onto a regular surface can make the speckle on an irregular surface include at least one speckle, preferably multiple speckles. The object may be associated with at least a partially irregular surface. Therefore, a speckle image can include multiple speckles. For example, if patterned coherent electromagnetic radiation is at least partially projected onto a user's skin, multiple speckles are formed due to interference of the coherent electromagnetic radiation. Therefore, depending on the surface onto which the patterned coherent electromagnetic radiation is projected, the speckle can include zero, one, or more speckles. Skin may have an irregular surface. Therefore, projecting patterned coherent electromagnetic radiation may result in the formation of speckle within one or more speckles. Projecting coherent electromagnetic radiation onto an irregular surface results in the formation of speckle. Therefore, the speckle can include one or more speckles. The speckle may have a diameter between 0.5 mm and 5 cm, preferably between 0.6 mm and 4 cm, more preferably between 0.7 mm and 3 cm, and most preferably between 0.4 cm and 2 cm.

[0072] As used herein, the term "speckle image" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, an image showing multiple speckles. A speckle image can show multiple speckles. A speckle image can include an image showing at least a portion of a user, particularly the user's face, when irradiated with coherent electromagnetic radiation, particularly over a corresponding region of interest included in the image. A speckle image can be generated when a user is irradiated with coherent electromagnetic radiation associated with wavelengths between 850 nm and 1400 nm. A speckle image can show a speckle pattern. A speckle pattern can specify the distribution of speckles. A speckle image can indicate the spatial extent of the speckles. A speckle image can be suitable for determining a measure of surface roughness. A speckle image can be generated using at least one camera with a detector, such as a TOF system. To generate a speckle image, a user can be irradiated by an illumination source.

[0073] As used herein, the term "speckle" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, optical phenomena caused by the interference of coherent electromagnetic radiation due to the irregularity or irregularity of surfaces. Speckle can manifest as variations in contrast in an image (such as a speckle image).

[0074] As used herein, the term "speckle pattern" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, the distribution of multiple speckles. The distribution of multiple speckles may refer to the spatial distribution of at least one speckle among multiple speckles and / or the spatial distribution of at least two speckles relative to each other. The spatial distribution of at least one speckle among multiple speckles may refer to and / or specify the spatial extent of that at least one speckle among multiple speckles. The spatial distribution of at least two speckles among multiple speckles may refer to and / or specify the spatial extent of a first speckle among at least two speckles relative to a second speckle among at least two speckles, and / or the distance between the first speckle among at least two speckles and the second speckle among at least two speckles.

[0075] As mentioned above, since speckle patterns may be caused by surface irregularities, they reflect the surface roughness. Therefore, determining surface roughness metrics based on speckle patterns in speckle images utilizes the relationship between speckle distribution and surface roughness. This provides a low-cost, efficient, and readily available solution for surface roughness assessment.

[0076] The generation of speckle images can be initiated by user action or can be initiated automatically, for example, when the presence of a user is automatically detected within the camera's field of view and / or a predetermined area of ​​the field of view. As used herein, the term "field of view" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, the angular range of the observable world and / or at least one scene that can be captured or viewed by an optical system (such as an image generation unit). The field of view is typically expressed in degrees and / or radians, and exemplaryly may represent the total angle spanned by the image and / or the visible area.

[0077] For example, a camera may include at least one image sensor and at least one additional optical element. For example, the additional optical element may be at least one lens. A lens may refer to an optical element adapted to influence the beam spread associated with coherent electromagnetic radiation. For example, the additional optical element may be at least one polarizing mirror. For example, a camera may include at least one image sensor, at least one lens, and at least one polarizing mirror. A polarizing mirror may refer to an optical element adapted to select electromagnetic radiation according to its polarization. In particular, a polarizing mirror may be an optical element adapted to select coherent electromagnetic radiation according to its polarization. Thus, a portion of the electromagnetic radiation, especially coherent electromagnetic radiation, can pass through the polarizing mirror, while the remaining electromagnetic radiation, especially coherent electromagnetic radiation, can be at least partially deflected away and / or at least partially absorbed. Since coherent electromagnetic radiation associated with wavelengths between 850 nm and 1400 nm penetrates deep into the skin, a portion of the information received from the light reflected from the skin includes information unrelated to surface roughness, which interferes with the measurement of surface roughness. To increase the signal-to-noise ratio, a polarizing mirror can be used. Typically, the coherent electromagnetic radiation reflected from the surface of an object is polarized differently from the light reflected from deeper layers of human skin. Therefore, polarizing mirrors enable the selection of desired signals from unwanted signals.

[0078] For example, the distance between the user and the camera used to generate the speckle image is between 10 cm and 1.5 m, and / or the distance between the user and the illumination source used to illuminate the user is between 10 cm and 1.5 m. Preferably, the distance between the user and the camera can be between 20 cm and 1.2 m. Preferably, the distance between the user and the illumination source can be between 20 cm and 1.2 m. Adjusting the distance between the object and the camera ensures that a speckle image of sufficiently good quality is generated. Therefore, the distances specified above enable accurate and reliable determination of surface roughness. This is particularly important in a non-static context where the user can operate the device themselves.

[0079] For example, a speckle image can show a user under coherent electromagnetic radiation and determine the surface roughness of the user's skin. Preferably, the user may have generated the speckle image and / or initiated its generation. Preferably, the speckle image can be initiated by the user-operated application of a mobile electronic device. By doing so, the user can decide for themselves when to determine the surface roughness of their skin. This allows non-expert users to determine surface roughness, and the measurement can be performed in a more natural and less artificial context. Consequently, surface roughness can be assessed more realistically, which in turn provides a more accurate measure of surface roughness. For example, depending on a person's activities, the surface roughness of the skin may vary throughout the day. Exercise can affect surface roughness and cause skin creaming. This effect can be verified using the methods and systems described herein.

[0080] For example, a speckle image can be associated with a resolution of less than 5 megapixels. Preferably, the speckle image can be associated with a resolution of less than 3 megapixels, more preferably less than 2.5 megapixels, and most preferably less than 2 megapixels. Such a speckle image can be generated using readily available, small, and inexpensive smartphone cameras. Furthermore, the storage and processing power required to evaluate surface roughness is relatively small. Therefore, the low resolution of the speckle image used to evaluate surface roughness enables the use of mobile electronic devices, particularly devices like smartphones or wearable devices, which have strictly limited size, memory, and processing power.

[0081] This method can further include reducing the speckle image to a predefined size before determining the surface roughness measure. Reducing the speckle image to the predefined size can be based on applying one or more image enhancement techniques. Reducing the speckle image to the predefined size can include: selecting a speckle image region of a predefined size, and segmenting the speckle image into speckle image regions of predefined sizes. The speckle image regions of predefined sizes can be associated with a living organism, such as a human being, and particularly with the skin of a living organism, such as human skin. The portion of the image outside the speckle image regions of predefined sizes may be associated with the background and / or may be unrelated to a living organism, such as a human being. By doing so, the amount of data that needs to be processed is reduced, which reduces the time required to determine the surface roughness, or allows for the need for less storage and processors. Furthermore, the image portions useful for analysis are selected. Therefore, reducing the size allows for the ignoring of portions of the speckle image that are unrelated to the object or living organism (such as a human). Thus, the surface roughness measure can be easily determined, and for analysis purposes, obstructive portions irrelevant to the user are ignored.

[0082] For example, image enhancement techniques may include at least one of the following: scaling, cropping, rotating, blurring, distorting, shearing, resizing, folding, changing contrast, changing brightness, adding noise, multiplying by at least a portion of pixel values, filtering, adjusting color, applying convolution, imprinting, sharpening, flipping, and averaging pixel values.

[0083] For example, the method may further include reducing the speckle image to a predefined size based on detecting a user in the speckle image. Specifically, the speckle image may be reduced to a predefined size based on detecting a user in the speckle image before determining the surface roughness metric. Specifically, reducing the speckle image to a predefined size based on detecting a user in the speckle image may include: detecting the user's contour (e.g., detecting the contour of the user's face), and reducing the speckle image to a region associated with the user (particularly a region associated with the user's face). Preferably, the region associated with the user may be within the contour of the user (particularly the user's) contour and / or the contour of the user's face.

[0084] The method may further include receiving at least one floodlight image. The floodlight image may include an image showing the user, particularly the user's face, while the user is being illuminated by the floodlight. The floodlight image can be generated by imaging and / or recording the light reflected from the user illuminated by the floodlight. The floodlight image showing the user may include at least a portion of the floodlight on at least a part of the user.

[0085] A floodlight image can reveal the contours of a user. The user's contours can be detected based on the floodlight image. Preferably, the user's contours can be detected by providing the floodlight image to an object detection data-driven model, particularly a user detection model, wherein the object detection data-driven model can be parameterized and / or trained based on a training dataset to receive the floodlight image and provide indications of the user's contours. The training dataset may include the floodlight image and indications of the contours of objects and / or people. The contour indications may include multiple points indicating the location of specific landmarks associated with the user. For example, where a speckle image can be associated with a user's face, the user's face can be detected based on the contours, where the contours may indicate facial landmarks such as the nose point, lip corners, or eyebrow tails.

[0086] As used herein, the term "surface roughness" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, the characteristics of a surface in relation to a user. In particular, surface roughness may characterize the lateral and / or vertical extent of a surface feature. Surface roughness can be evaluated based on surface roughness metrics. Surface roughness metrics can quantify surface roughness.

[0087] As used herein, the term "surface feature" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, any structure of any shape associated with a surface, particularly a user's surface. In particular, a surface feature may refer to a substructure of a surface associated with a user. A surface may include multiple surface features. For example, a ridge or depression may be a surface feature. Preferably, a surface feature may refer to a portion of a surface associated with an angle not equal to 90° relative to the surface normal.

[0088] As used herein, the term "surface roughness metric" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, a measure suitable for quantifying surface roughness. A surface roughness metric may be associated with a speckle pattern. For example, a surface roughness metric may include at least one of the following: fractal dimension, speckle size, speckle contrast, speckle modulation, roughness index, standard deviation of the height associated with a surface feature, lateral correlation length, mean median height, root mean square height, or a combination thereof. Preferably, a surface roughness metric may be suitable for describing vertical and lateral surface features. A surface roughness metric may include values ​​associated with the surface roughness metric. A surface roughness metric may refer to a term used to measure a quantity of surface roughness and / or a value associated with a quantity used to measure surface roughness. Determining a surface roughness metric based on a speckle image may mean determining the surface roughness metric based on the speckle pattern in the speckle image.

[0089] Surface roughness metrics can be determined based on speckle images by providing them to the model and receiving surface roughness metrics from the model. For example, the model can be adapted to determine the output based on the input. In particular, the model can be adapted to determine surface roughness metrics based on speckle images, preferably based on the received speckle images.

[0090] For example, a model can be, or may include, one or more of a physical model, a data-driven model, or a hybrid model. A hybrid model can be a model that includes at least one data-driven model with physical or statistical adjustments and model parameters. Statistical or physical adjustments can be introduced to improve the quality of the results because these adjustments provide a systematic relationship between empiricism and theory. For example, a data-driven model can represent the correlation between surface roughness metrics and speckle images. A data-driven model can obtain the correlation between surface roughness metrics and speckle images based on a training dataset that includes multiple speckle images and multiple surface roughness metrics. For example, a data-driven model can be parameterized based on a training dataset to receive speckle images and provide surface roughness metrics. A data-driven model can be trained based on a training dataset. A training dataset can include at least one speckle image and at least one corresponding surface roughness metric. A training dataset can include multiple speckle images and multiple surface roughness metrics. Training the model can include parameterizing the model. A data-driven model can be parameterized and / or trained to provide surface roughness metrics based on speckle images, particularly received speckle images. Determining a surface roughness measure based on a speckle image can include: providing the speckle image to a data-driven model, and receiving the surface roughness measure from the data-driven model. Providing a surface roughness measure based on a speckle image can include: mapping the speckle image to a surface roughness measure. The data-driven model can be parameterized and / or trained to receive the speckle image. The data-driven model can receive the speckle image at an input layer. The term "training" can also refer to learning. This term can specifically refer to, but is not limited to, the process of constructing a data-driven model, and in particular, determining and / or updating the parameters of the data-driven model. Updating the parameters of the data-driven model can also be referred to as retraining. Training as discussed herein can include retraining. During training, the data-driven model can be adjusted to achieve a best fit with the training data, for example, to best fit at least one input value to at least one desired output value. For example, if the neural network is a feedforward neural network (such as a CNN), a backpropagation algorithm can be applied to train the neural network. In the case of an RNN, a gradient descent algorithm or a backpropagation algorithm over time can be used to achieve the training objective. Training a data-driven model can include or can refer to calibrating a model, but is not limited to calibrating a model.

[0091] For example, a physical model can reflect physical phenomena mathematically, including, for instance, a first-principles model. A physical model can include a set of equations describing the interaction between an object and coherent electromagnetic radiation, thereby generating a measure of surface roughness. The physical model can be based on at least one of the following: fractal dimension, speckle size, speckle contrast, speckle modulation, roughness index, standard deviation of the height associated with a surface feature, lateral correlation length, mean median height, root mean square height, or a combination thereof. Specifically, a physical model can include one or more equations relating a speckle image to a measure of surface roughness, based on equations related to fractal dimension, speckle size, speckle contrast, speckle modulation, roughness index, standard deviation of the height associated with a surface feature, lateral correlation length, mean median height, root mean square height, or a combination thereof.

[0092] For example, the fractal dimension can be determined based on the Fourier transform and / or the inverse Fourier transform of the speckle image. For instance, the fractal dimension can be determined based on the slope of a linear function fitted to a double log-log plot of power spectral density versus frequency obtained through the Fourier transform. The speckle size can refer to the spatial extent of one or more specks. Where the speckle size can refer to the spatial extent of more than one speckle, the speckle size can be determined based on the average of more than one speckle size and / or a weighted average of more than one speckle size. The speckle contrast can refer to a measure of the standard deviation of the average intensity of at least a portion of the speckle image relative to at least a portion of the average intensity of the speckle image. The speckle modulation can refer to a measure of the intensity fluctuations associated with speckles in at least a portion of the speckle image. The roughness index, the standard deviation of the height associated with surface features, the lateral correlation length, or a combination thereof can be determined based on an autocorrelation function associated with a double log-log plot of power spectral density versus frequency obtained through the Fourier transform.

[0093] For example, determining a surface roughness metric based on a speckle image can include determining the surface roughness metric based on a speckle pattern. For example, determining surface roughness based on a speckle pattern can include determining surface roughness based on the distribution of multiple specks in a speckle image. Determining a surface roughness metric based on the distribution of multiple specks in a speckle image can refer to determining the distribution of multiple specks in a speckle image. Determining the speckle distribution can include determining at least one of the following: the fractal dimension associated with the speckle image, the speckle size associated with the speckle image, the speckle contrast associated with the speckle image, the speckle modulation associated with the speckle image, the roughness index associated with the speckle image, the standard deviation of the height associated with the surface features associated with the speckle image, the lateral correlation length associated with the speckle image, the mean median height associated with the speckle image, the root mean square height associated with the speckle image, or a combination thereof.

[0094] Alternatively or concurrently, determining a surface roughness metric may include determining at least one of the following: fractal dimension associated with a speckle image, speckle size associated with a speckle image, speckle contrast associated with a speckle image, speckle modulation associated with a speckle image, roughness index associated with a speckle image, standard deviation of height associated with surface features associated with a speckle image, lateral correlation length associated with a speckle image, mean median height associated with a speckle image, root mean square height associated with a speckle image, or a combination thereof.

[0095] For example, determining a surface roughness metric can be based on the distribution of speckles in a speckle image. Determining a surface roughness metric based on the distribution of speckles in a speckle image can include determining at least one of the following: speckle size distribution, power spectral density associated with the speckle image, fractal dimension associated with the speckle image, speckle contrast, speckle modulation, or a combination thereof.

[0096] Alternatively or alternatively, determining surface roughness metrics based on the distribution of speckles in a speckle image may include providing the speckle image to a model, particularly a data-driven model, wherein the data-driven model may be parameterized and / or trained based on a training dataset comprising one or more speckle images and one or more corresponding surface roughness metrics.

[0097] For example, a surface roughness metric can be determined based on a speckle image by providing the speckle image to the model and receiving the surface roughness metric from the model. The model can be a data-driven model and can be parameterized and / or trained based on a training dataset that includes multiple speckle images and corresponding surface roughness metrics or indicators of surface roughness metrics. Alternatively, the model can be a physical model.

[0098] For example, the method may further include generating a partial speckle image. A partial speckle image can refer to a portion of the image generated based on the speckle image. A partial speckle image can be generated by applying one or more image enhancement techniques to the speckle image.

[0099] For example, the method may further include generating a first speckle image and a second speckle image. The speckle image may include a first speckle image and a second speckle image. The first speckle image may refer to a first portion of the speckle image. The second speckle image may refer to a second portion of the speckle image. Preferably, the first speckle image and the second speckle image may be different from each other. In particular, the first speckle image and the second speckle image may be non-overlapping. The first speckle image and the second speckle image can be generated by applying one or more image enhancement techniques to the speckle image. Determining a surface roughness measure based on the speckle image may include: determining a first surface roughness measure based on the first speckle image, and determining a second surface roughness measure based on the second speckle image. Providing a surface roughness measure may include providing a first surface roughness measure and a second surface roughness measure. In particular, the first surface roughness measure and the second surface roughness measure may be provided together. Preferably, the first surface roughness measure and the second surface roughness measure may be provided in a surface roughness measure map indicating the spatial distribution of the surface roughness measure. For example, a surface roughness metric map can indicate a first surface roughness metric associated with a first region in the surface roughness metric map, and a second surface roughness metric map can indicate a second surface roughness metric associated with a second region in the surface roughness metric map. In particular, the surface roughness metric map can resemble a heatmap, wherein the surface roughness metric can be plotted for the region associated with the corresponding surface roughness metric.

[0100] Surface roughness metrics can be determined by using at least one processor.

[0101] For example, the method may include determining whether a surface roughness metric corresponds to a human surface roughness metric. For example, the method may include determining whether a surface roughness metric corresponds to a specific user surface roughness metric. Determining whether a surface roughness metric corresponds to a human and / or a specific user surface roughness metric may include comparing the surface roughness metric to at least one predefined or predetermined value range of the surface roughness metric (e.g., stored in at least one database, such as on a device, or stored in a remote database, such as in the cloud). If the determined surface roughness metric is at least within the tolerance of the redefined or predetermined value range of the surface roughness metric, the user is authenticated; otherwise, authentication fails. For example, the surface roughness metric may be human skin roughness. If the determined human skin roughness is in the range of 10 µm to 150 µm, the object is considered human. However, other ranges are also possible.

[0102] This method can be computer-implemented. As used herein, the term "computer-implemented" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, methods involving at least one computer and / or at least one computer network. The computer and / or computer network may include at least one processor configured to perform at least one method step of the method according to the invention. Specifically, each of these method steps is performed via a computer and / or computer network. The method can be performed entirely automatically, specifically without user interaction. For example, irradiation and / or image generation can be triggered and / or performed using at least one processor.

[0103] A single processing device may be configured to exclusively execute at least one computer program, particularly at least one line of computer program code configured to execute at least one algorithm, as used in at least one embodiment of the method according to the invention. Hereinafter, the computer program executing on the single processing device may include all instructions that cause the computer to perform the method. Alternatively or additionally, at least one method step may be performed using at least one remote device, particularly selected from at least one of a server or a cloud server, especially when the device and the remote device may be part of a computer network. In this case, the computer program may include at least one remote component to be executed by at least one remote processing device to perform at least one method step. Further, the computer program may include at least one interface configured to forward data to and / or receive data from at least one remote component of the computer program.

[0104] On the other hand, a system for determining whether an object corresponds to a living organism is disclosed. The system includes:

[0105] - At least one time-of-flight system, comprising at least one transmitting device for generating a light beam pattern to illuminate the object, wherein the transmitting device includes at least one light emitter array, and wherein the time-of-flight system further includes at least one detector configured to generate at least one patterned image of the object while the object is illuminated by the light beam pattern.

[0106] - At least one processor configured to: extract live data from the pattern image using beam profiling analysis, and determine, based on the live data, whether the object corresponds to a living organism.

[0107] As used herein, the term "processor" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, any logical circuit system configured to perform basic operations of a computer or system, and / or generally to a device configured to perform computations or logical operations. In particular, a processor or computer processor may be configured to process the basic instructions that drive a computer or system. A processor may be a semiconductor-based processor, a quantum processor, or any other type of processor configured to process instructions. As an example, a processor may be or may include a Central Processing Unit ("CPU"). A processor may be a Graphics Processing Unit ("GPU"), a Tensor Processing Unit ("TPU"), a Complex Instruction Set Computing Microprocessor ("CISC"), a Reduced Instruction Set Computing ("RISC") microprocessor, a Very Long Instruction Word ("VLIW") microprocessor, or a processor implementing other instruction sets or multiple processors implementing combinations of instruction sets. The processing device can also be one or more dedicated processing devices, such as application-specific integrated circuits (“ASICs”), field-programmable gate arrays (“FPGAs”), complex programmable logic devices (“CPLDs”), digital signal processors (“DSPs”), network processors, etc. The methods, systems, and devices described herein can be implemented as software in a DSP, microcontroller, or any other auxiliary processor, or as hardware circuitry within an ASIC, CPLD, or FPGA. It should be understood that the term processor can also refer to one or more processing devices, such as a distributed processing device system located on multiple computer systems (e.g., cloud computing), and is not limited to a single device unless otherwise stated. A processor can also be an interface to a remote computer system, such as a cloud service. A processor can include or be a secure isolated zone processor (SEP). An SEP can be a secure circuit configured for processing images. A “secure circuit” is circuitry that protects isolated internal resources from direct access by external circuitry. A processor can be an image signal processor (ISP) and can include circuitry systems suitable for processing images, particularly images containing personal and / or confidential information.

[0108] The system can be configured to perform a method for determining whether an object corresponds to a living organism according to the present invention (e.g., according to one or more embodiments given above or further detailed below). For details, options, and definitions, reference can be made to the method for determining whether an object corresponds to a living organism as discussed above, and the method and apparatus described below.

[0109] On the other hand, the present invention discloses a method for authenticating a user of a device, particularly for performing at least one operation requiring authentication on the device. These method steps may be performed in a given order or in a different order. Further, one or more additional method steps not listed may be present. Further, one, more than one, or even all of the method steps may be performed repeatedly. For details, options, and definitions, refer to the methods and systems for determining whether an object corresponds to a living organism as discussed above, and the methods and apparatus described below.

[0110] As used herein, the term "user" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, a person who intends to use the device and / or uses the device.

[0111] The device can be selected from the group consisting of: television equipment; game consoles; personal computers; mobile devices, especially mobile phones, and / or smartphones, and / or tablet computers, and / or laptop computers, and / or tablet computers, and / or virtual reality devices, and / or wearable devices such as smartwatches; or other types of portable computers.

[0112] As used herein, the term "authentication" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, verifying the identity of a user. Specifically, authentication may include distinguishing a user from other humans or objects, particularly distinguishing authorized access from unauthorized access.

[0113] Authentication may include verifying the identity of the corresponding user and / or assigning an identity to the user. Authentication may include generating and / or providing identity information, such as providing it to other devices or units (e.g., to at least one authorized unit) to authorize access to the device. The identity information can be proven through authentication. For example, the identity information may be and / or may include at least one identity token. In the event of successful authentication, the facial image recorded by the imaging detector (such as the imaging detector of a TOF system, or another imaging detector) can be verified as an image of the user's face, and / or the user's identity is verified.

[0114] Authentication can be performed using at least one authentication process. The authentication process may include multiple steps, such as at least one face detection (e.g., on at least one floodlight image, as will be described in more detail below), and at least one recognition step, wherein an identity is assigned to the detected face and / or at least one identity check and / or verification of the user's identity is performed.

[0115] Authentication can be and / or may include biometric authentication. As used herein, the term "biometric authentication" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, authentication using at least one biometric identifier (e.g., a unique, measurable characteristic used to identify and describe an individual). A biometric identifier may be a physiological characteristic.

[0116] The method includes receiving a request for accessing at least one resource associated with the device and performing at least one authentication process. The authentication process includes...

[0117] a. Determine the user's distance information by using a time-of-flight system, and determine whether the user is within or outside the scope of the authentication process by comparing the distance information with the working area;

[0118] b. If it is determined that the user is within the scope of the work, allow the user to access the resource; otherwise, if it is determined that the user is outside the scope of the work, deny the user access to the resource.

[0119] The device can be selected from the group consisting of: television equipment; game consoles; personal computers; mobile devices, especially mobile phones, and / or smartphones, and / or tablet computers, and / or laptop computers, and / or tablet computers, and / or virtual reality devices, and / or wearable devices such as smartwatches; or other types of portable computers.

[0120] As used herein, the term "access" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, accessing and / or using one or more functions associated with a device. As used herein, the term "function associated with a device" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, any function, such as accessing at least one element and / or at least one resource of the device or associated with it. Functions requiring user authentication may be predefined. One or more functions associated with a device may include unlocking the device and / or accessing an application preferably associated with the device and / or accessing a portion of an application preferably associated with the device. For example, the function may include accessing content of the device, such as content stored in the device's database and / or content that can be retrieved by the device. In embodiments, allowing a user to access a resource may include allowing a user to perform at least one operation with the device and / or system. A resource may be a device, a system, a function of the device, a function of the system, and / or an entity. Additionally and / or alternatively, allowing a user to access a resource may include allowing a user to access an entity. An entity may be a physical entity and / or a virtual entity. Virtual entities can be, for example, databases. Physical entities can be restricted areas. Restricted areas can be one of the following: secure areas, rooms, apartments, vehicles, portions of the examples mentioned earlier, etc. Devices may be locked and can only be unlocked by authorized users.

[0121] The device may further include at least one communication interface, such as a user interface, configured to receive requests for access to at least one resource associated with the device, particularly for performing at least one operation requiring authentication on the device. As used herein, the term "request for access" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, at least one action and / or instance of requesting access. As used herein, the term "receiving a request" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, a process of obtaining a request, for example, from a data source and / or a user interface. Receiving may be fully or partially automated. Receiving a request for access to one or more functions associated with the device can be performed using at least one communication interface. Receiving may include, for example, receiving at least one user input via at least one user interface (e.g., the device's display), and / or receiving requests from remote devices and / or the cloud (e.g., via device communication, such as via the Internet). For example, a request can be generated or triggered by at least one user input (e.g., via entering a security number or other unlocking action by the user), and / or can be sent from a remote device and / or the cloud (e.g., via a connected account).

[0122] The authentication process can be executed using at least one authentication unit configured to perform at least one authentication procedure for a user. An authentication unit may include at least one processor. Execution of the authentication process can be triggered and / or started by receiving a request.

[0123] The certification process can include multiple steps.

[0124] For example, the authentication process may include performing at least one face detection step. The face detection step may include analyzing at least one image of the user, generated, for example, by a detector from a Time-of-Flight (TOF) system or another camera. The image may be a floodlight image.

[0125] The authentication process may further include generating at least one floodlight image showing the associated user while the user is being illuminated by the floodlight, and determining, based on the floodlight image, whether the user's identity corresponds to a verified identity. The authentication process may include: allowing the user to access the resource if the user's identity corresponds to a verified identity, and otherwise denying the user access to the resource if the user's identity does not correspond to a verified identity.

[0126] The method may include:

[0127] -Utilize floodlight illumination on users by using at least one floodlight source;

[0128] - Capture at least one floodlight image by using at least one image generation unit (such as a detector of a TOF system or another camera).

[0129] As used herein, the term "floodlight source" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, at least one arbitrary device configured to provide substantially continuous spatial illumination. As used herein, the term "floodlight" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, substantially continuous spatial illumination, particularly diffuse and / or uniform illumination. Floodlight has wavelengths in the infrared range, particularly in the near-infrared range. A floodlight source may include at least one LED or at least one VCSEL, preferably multiple VCSELs. As used herein, the term "substantially continuous spatial illumination" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, uniform spatial illumination, wherein non-uniform areas are possible. The area illuminated from the floodlight source, for example, covering a user, a portion of a user, and / or the user's face, may be continuous. Power may be distributed across the entire illumination field. In contrast, illumination provided by a light pattern can include at least two consecutive regions, particularly multiple consecutive regions, and / or the power can be concentrated in a smaller area of ​​the illumination field (compared to the entire illumination field). Infrared flood illumination can be suitable for illuminating consecutive regions, particularly a single consecutive region. Infrared pattern illumination can be suitable for illuminating at least two consecutive regions.

[0130] As used herein, the term "floodlight image" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, an image generated by an image generation unit when an illumination source emits infrared floodlight (e.g., on an object and / or a user). A floodlight image may include an image showing a user, particularly the user's face, while the user is being illuminated using floodlight. A floodlight image can be generated by imaging and / or recording light reflected from an object and / or user illuminated by floodlight. A floodlight image showing a user may include at least a portion of the floodlight on at least a part of the user. For example, illumination and imaging of the floodlight source may be synchronized, for example, by using at least one control unit.

[0131] The face detection step may include analyzing a floodlight image. For example, an authentication process may include performing at least one face detection using a floodlight image. Face detection can be performed locally on the device. However, face recognition (i.e., assigning an identity to a detected face) can be performed remotely, for example, in the cloud, especially when identification rather than just verification is required. User templates can be stored at a remote device, such as in the cloud, and do not need to be stored locally. This can be advantageous from a storage and security perspective.

[0132] The authentication process may include identifying a user based on a floodlight image. As used herein, the term "identification" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, assigning identity to a detected face and / or at least one identity check and / or verification of the user's identity. Therefore, in particular, the authentication unit may forward data to a remote device. Alternatively or additionally, the authentication unit may perform user identification based on a floodlight image, particularly by running an appropriate computer program with corresponding functionality.

[0133] Recognition may include assigning an identity to a detected face and / or verifying the user's identity. Recognition may include performing facial verification on the imaged face to confirm whether it is the user's face. Identifying the user may include matching a floodlight image (e.g., showing the outline of parts of the user, particularly parts of the user's face) against a template. To match the floodlight image with the template, the similarity between at least one image feature vector obtained from the floodlight image and at least one template feature vector may be considered and / or evaluated. This template vector may be obtained from a template image.

[0134] The template image can be generated during the registration process. As used herein, the term "registration process" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, at least one step of registration, particularly registration with the service. During the registration process, the template image can be generated under secure conditions, in a manner that ensures the generated template image is presented to the user. The registration process may include at least one of the following steps: capturing the template image; recording personal data, etc.

[0135] Face detection may include analyzing floodlight images. Specifically, the analysis of floodlight images may include using at least one image recognition technique, particularly face recognition technology. Image recognition technology includes at least one process for identifying a user in an image. Image recognition may include using at least one technique selected from the following: color-based image recognition, for example using features such as template matching; segmentation and / or connected component (blob) analysis, for example using size or shape; machine learning and / or deep learning, for example using at least one convolutional neural network.

[0136] For example, authentication may include identifying a user. Identification may include assigning an identity to a detected face and / or at least one identity check and / or verification of the user's identity. Identification may include performing facial verification on the imaged face to confirm whether it is the user's face. Identifying a user may include matching a flood image (e.g., showing the outlines of various parts of the user, particularly the outlines of various parts of the user's face) with a template (e.g., a template image generated during registration). Identifying a user may include determining whether the imaged face is the user's face, and in particular determining whether the imaged face corresponds to at least one image of the user's face stored in at least one memory of a device, for example. If the flood image can match the image template, authentication may be successful. If the flood image cannot match the image template, authentication may be unsuccessful.

[0137] As used herein, the term "memory" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. Specifically, the term may refer to, but is not limited to, at least one electronic storage space configured to store data, instructions, and programs. The stored data, instructions, and / or programs can be forwarded to a processor for processing. Memory can be or may include at least one of the following: random access memory; read-only memory; cache memory; hard disk drive; solid-state drive; virtual memory.

[0138] To determine whether a user's identity corresponds to a verified identity based on a floodlight image, the similarity between at least one image feature vector obtained from the floodlight image and at least one template feature vector can be considered. This template vector can be obtained from a template image, which can be generated during the registration process.

[0139] For example, user identification may include determining multiple facial features. Analysis may include comparing the determined facial features with template features, specifically performing a matching process. Template features may be features extracted from at least one template. The template may be or may include at least one image generated during registration (e.g., when initializing the device). The template may be an image of an authorized user. Template features and / or facial features may include vectors. Feature matching may include determining the distance between vectors. User identification may include comparing the distance between vectors with at least one predefined limit. If the distance is at least within tolerance and ≤ the predefined limit, the user can be successfully identified. Otherwise, the user may be rejected and / or refused.

[0140] Analysis of a flood image may further include one or more of the following: filtering; selecting at least one region of interest; forming a difference image between the flood image and at least one offset; inverting the flood image; background correction; decomposing into color channels; decomposing into hue, saturation, and brightness channels; frequency decomposition; singular value decomposition; applying a Canny edge detector; applying a Laplacian Gaussian filter; applying a difference Gaussian filter; applying the Sobel operator; applying the Laplacian operator; applying the Scharr operator; applying the Prewitt operator; applying the Roberts operator; applying the Kirsch operator; applying a high-pass filter; applying a low-pass filter; applying a Fourier transform; applying the Radon transform; applying the Hough transform; applying the wavelet transform; thresholding; and creating a binary image. The region of interest may be manually determined by the user or automatically determined, for example, by identifying the user within the image.

[0141] For example, image recognition may include a trained model using at least one model, particularly one including at least one face recognition model. Analysis of floodlight images can be performed using a face recognition system such as FaceNet, as described, for example, in Florian Schroff, Dmitry Kalenichenko, and James Philbin, “FaceNet: A Unified Embedding for Face Recognition and Clustering,” arXiv:1503.03832. The trained model may include at least one convolutional neural network. For example, a convolutional neural network may be designed as described in the following literature: MD Zeiler and R. Fergus, “Visualizing and understanding convolutional networks,” CoRR, abs / 1311.2901, 2013; or C. Szegedy et al., “Going deeper with convolutions,” CoRR, abs / 1409.4842, 2014. For more details on convolutional neural networks for face recognition systems, please refer to: Florian Schroff, Dmitry Kalenichenko, and James Philbin, “FaceNet: A Unified Embedding for Face Recognition and Clustering,” arXiv:1503.03832. Labeled image data from image databases can be used as training data.Specifically, labeled faces can be used from one or more of the following sources: GB Huang, M. Ramesh, T. Berg, and E. Learned-Miller, “Labeled faces in the wild: A database for studying face recognition in unconstrained environments,” Technical Report 07-49, University of Massachusetts Amherst, October 2007; the YouTube® Faces database as described in L. Wolf, T. Hassner, and I. Maoz, “Face recognition in unconstrained videos with matched background similarity,” IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2011; or the Google® Facial Expression Comparison Dataset. Training of convolutional neural networks can be described as in "FaceNet: A Unified Embedding for Face Recognition and Clustering" by Florian Schroff, Dmitry Kalenichenko, and James Philbin, arXiv:1503.03832.

[0142] This method may include determining the distance of an object to the detector of the TOF system using a Time-of-Flight (TOF) system. The method may include determining whether the distance between the object and the detector of the TOF system is within the operating range, and allowing user access to resources in response to determining that the distance is within the operating range. This invention allows for consideration of distance information obtained by the TOF system for the authentication process. The detector of the TOF system is adapted to determine the distance and generate an image for authentication. Therefore, advantageously, space in the device can be saved by using a TOF system for authentication.

[0143] Distance is also important for facial authentication because the facial recognition model is associated with a working range. The working range can specify the range of distances between a user's face and a detector for generating an image of the user, on which the facial recognition model operates and / or is trained. In embodiments, the working range can specify at least one upper boundary and / or at least one lower boundary of the distance of an object from the detector and / or illumination source. The working range can be associated with an authentication process. The working range can include at least one value. This value can be numerical, particularly a positive value. An indication of the working range can be received, particularly before determining whether the distance is within or outside the working range of the authentication process. The indication of the working range can be adapted to determine whether the distance is within or outside the working range of the authentication process. The indication of the working range can be adapted to compare the distance with the working range.

[0144] Distance information can be determined directly in the detector of the TOF system. Determining distance information may include considering, for example, calibration data stored in the detector's memory. The detector may include at least one communication interface configured to provide the user's image to another unit for further analysis and / or to provide distance information to an authentication unit to check whether the distance is within or outside the operating range.

[0145] As used herein, the term "communication interface" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to a specific or custom meaning. Specifically, the term may refer to, but is not limited to, an article or element forming a boundary configured for transmitting information. In particular, a communication interface may be configured to transmit information from a computing device (e.g., a computer), such as to send or output information to, for example, another device. Additionally or alternatively, a communication interface may be configured to transmit information to a computing device, such as to a computer, such as to receive information. A communication interface may specifically provide a means for transmitting or exchanging information. In particular, a communication interface may provide a data transmission connection, such as Bluetooth, NFC, Ethernet, inductive coupling, etc. As an example, a communication interface may be or may include at least one port, including one or more of a network or internet port, a USB port, and a disk drive. A communication interface may be at least one network interface.

[0146] The authentication process may further include determining whether an object corresponds to a living organism by performing a method according to the invention (e.g., according to one or more embodiments given above or further detailed below) for determining whether an object corresponds to a living organism. The authentication process may include: allowing the user to access the resource if it is determined that the user's pattern image corresponds to a living organism, otherwise denying the user access to the resource if it is determined that the user's pattern image does not correspond to a living organism.

[0147] The authentication unit can be further configured to consider additional security features extracted from the pattern image. In particular, the authentication unit can be further configured to extract liveness data and / or consider liveness data extracted from the pattern image.

[0148] Therefore, specifically, the authentication unit can forward data to a remote device. Alternatively or additionally, the authentication unit can use patterned images to perform the extraction of liveness data, particularly by running an appropriate computer program with corresponding functionality. In particular, the authentication unit can treat liveness data as a parameter for verifying the authentication process, which can be robust to prevent deception by using the recorded user image.

[0149] The authentication unit can be configured to outsource at least one step of the authentication process (such as user identification) and / or at least one step of the verification process (such as consideration of material data) to a remote device, specifically a server and / or a cloud server. This device and the remote device can be part of a computer network, particularly the Internet. Thus, the device can function as a field device used by the user to generate data required in the authentication process and / or its verification. The device can transmit the generated data and / or data associated with intermediate steps of the authentication process and / or its verification to the remote device. In this scenario, the authentication unit can be and / or may include a connection interface configured to transmit information to the remote device. Data generated by the remote device used in the authentication process and / or its verification can be further transmitted to the device. This data can be received by the connection interface included in the device. The connection interface can be specifically configured to transmit or exchange information. In particular, the connection interface can provide a data transmission connection. As an example, the connection interface can be or may include at least one port, including one or more of a network or Internet port, a USB port, and a disk drive.

[0150] It is important to emphasize that data from a device can be transferred to a specific remote device based on at least one circumstance (e.g., date, day, load of a particular remote device, etc.). A field device may not be able to select a specific remote device. Conversely, another device may choose which specific remote device the data can be transferred to. The authentication process and / or the generation of verification data may involve several different entities using the remote device. At least one entity may generate intermediate data and transfer that intermediate data to at least one other entity.

[0151] The authentication unit can be further configured to determine distance information at multiple locations on the user's face using a Time-of-Flight (TOF) system and to determine a depth map. The determined depth map can be compared to a predetermined depth map of the user, for example, determined during the registration process. The authentication unit can be configured to authenticate the user if the determined depth map matches (in particular, at least within tolerance) the user's predetermined depth map. Otherwise, the user can be rejected.

[0152] Allowing a user to access resources may include authorizing the user. The device may include at least one authorization unit configured to allow the user to perform at least one operation on the device, such as unlocking the device, upon successful authentication, or to deny the user from performing at least one operation on the device if authentication fails. Thus, the user is aware of the authentication result. The authorization unit may be configured to allow or deny the user to perform at least one operation on the device requiring authentication based on material data and identification using floodlight images. The authorization unit may be configured to allow or deny the user access to one or more functions associated with the device, depending on authentication or denial. Allowing may include granting permission to access the one or more functions. The authorization unit may be configured to determine whether the user corresponds to an authorized user, wherein allowing or denying is further based on determining whether the user corresponds to an authorized user. As used herein, the term "authorization" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a special or custom meaning. The term may specifically refer to, but is not limited to, the process of assigning access rights to a user (particularly selective permission or selective restriction of access to the device and / or at least one resource of the device). The authorization unit may be configured for access control. As used herein, the term "authorization unit" is a broad term and will be given its common and conventional meaning to those skilled in the art, and is not limited to any particular or custom meaning. The term may specifically refer to, but is not limited to, a unit configured to authorize a user, such as a processor. An authorization unit may include at least one processor or may be designed as software or an application. The authorization unit and the authentication unit may be integrated, for example, by using the same processor. The authorization unit may be configured to allow a user to access, for example, one or more functions on the device, such as unlocking the device, upon successful authentication, or to deny the user access to, for example, one or more functions on the device, upon unsuccessful authentication.

[0153] For example, by using a user interface (such as the device's display), the device can be configured to display the results of authentication and / or authorization.

[0154] On the other hand, a system for authenticating users was disclosed. This system includes:

[0155] a. At least one time-of-flight system;

[0156] b. At least one processor; and

[0157] c. At least one memory configured to store instructions that, when executed by a processor, cause the system to perform a user method for authenticating a device according to the invention (e.g., according to one or more embodiments given above or further detailed below).

[0158] For details, options, and definitions, please refer to the methods described herein and the systems discussed above for determining whether an object corresponds to a living organism, as well as the methods and devices described below.

[0159] The system can be further configured to determine whether an object corresponds to a living organism. The time-of-flight system may include at least one emitting device for generating a beam pattern for illuminating the object. The emitting device may include at least one array of light emitters. The time-of-flight system may further include at least one detector configured to generate at least one pattern image of the object while it is being illuminated by the beam pattern. The processor can be configured to: extract liveness data from the pattern image using beam profiling analysis, and determine whether the object corresponds to a living organism based on the liveness data.

[0160] The system can be configured to perform a method according to the invention (such as one or more of the embodiments given above or further detailed below) for determining whether an object corresponds to a living organism.

[0161] This document further discloses and proposes a computer program comprising computer-executable instructions for performing one or more methods according to the invention in one or more embodiments of the examples included herein when the program is executed on a computer or computer network. Specifically, the computer program may be stored on a computer-readable data carrier and / or a computer-readable storage medium.

[0162] As used herein, the terms "computer-readable data carrier" and "computer-readable storage medium" specifically refer to non-transitory data storage devices, such as hardware storage media on which computer-executable instructions are stored. Computer-readable data carriers or storage media can specifically be or may include storage media such as random access memory (RAM) and / or read-only memory (ROM).

[0163] Therefore, specifically, one, more, or even all of the method steps i) to iii) and / or method steps a) to b) indicated above can be performed by using a computer or computer network, preferably by using a computer program.

[0164] This document further discloses and proposes a computer program product having program code means so that, when the program is executed on a computer or computer network, it performs one or more methods according to the invention as included in one or more embodiments herein. Specifically, the program code means may be stored on a computer-readable data carrier and / or a computer-readable storage medium.

[0165] This document further discloses and proposes a data carrier on which a data structure is stored, which, after being loaded into a computer or computer network (e.g., into the working memory or main memory of the computer or computer network), can perform one or both methods according to one or more embodiments disclosed herein.

[0166] This document further discloses and proposes a computer program product having program code means stored on a machine-readable medium to perform one or both methods according to one or more embodiments disclosed herein when the program is executed on a computer or computer network. As used herein, a computer program product refers to a program that is a tradable product. The product can generally exist in any format, such as in paper format, or on a computer-readable data carrier and / or computer-readable storage medium. Specifically, the computer program product can be distributed via a data network.

[0167] Finally, this document discloses and proposes a modulated data signal containing computer system or computer network readable instructions for performing one or both methods according to one or more embodiments disclosed herein.

[0168] Referring to the computer implementation aspects of the present invention, one or more, or even all, of the method steps in one or more of the methods disclosed in the embodiments herein can be performed using a computer or computer network. Therefore, typically, any of the method steps involving the provision and / or manipulation of data can be performed using a computer or computer network. Generally, these method steps can include any method steps, except for those that typically require manual work, such as providing samples and / or performing certain aspects of actual measurements.

[0169] Specifically, this article further discloses:

[0170] - A computer or computer network comprising at least one processor, wherein the processor is adapted to perform one or both methods according to one of the embodiments described in this specification.

[0171] - A computer-loadable data structure adapted to perform one or both of the methods described in this specification when the data structure is executed on a computer.

[0172] - A computer program, wherein the computer program is adapted, when executed on a computer, to perform one or both of the methods described in this specification, according to one of the embodiments described herein.

[0173] A computer program comprising program means for performing one or both of the methods described herein when the computer program is executed on a computer or a computer network.

[0174] - A computer program comprising program means according to a preceding embodiment, wherein the program means is stored on a computer-readable storage medium.

[0175] - A storage medium wherein a data structure is stored on the storage medium, and wherein the data structure is adapted to perform one or both methods according to one of the embodiments described in this specification after being loaded into the main storage device and / or working storage device of a computer or computer network.

[0176] - A computer program product having program code means, wherein the program code means may be stored or stored on a storage medium for performing one or both of the embodiments described herein when the program code means is executed on a computer or computer network.

[0177] On the other hand, the use of a time-of-flight system for authenticating users and / or determining whether an object corresponds to a living organism is disclosed. A time-of-flight system may be included in systems for authenticating users and / or determining whether an object corresponds to a living organism according to the present invention (such as one or more of the embodiments given above or further detailed below).

[0178] On the other hand, the images generated by the detectors of the time-of-flight system are used for authenticating users and / or determining whether an object corresponds to a living organism. The detectors of the time-of-flight system may be included in systems for authenticating users and / or determining whether an object corresponds to a living organism, according to one or more embodiments given above or further detailed below, as described in the invention.

[0179] As used herein, the terms “have,” “include,” or “contain,” or any of their grammatical variations, are used in a non-exclusive manner. Thus, these terms can refer either to a situation where no other features exist in the entity described in the context besides those introduced by these terms, or to a situation where one or more other features exist. For example, the statements “A has B,” “A includes B,” and “A contains B” can refer either to a situation where no other elements exist in A besides B (i.e., A consists solely of B), or to a situation where entity A contains one or more other elements besides B (such as element C, elements C and D, or even other elements).

[0180] Furthermore, it should be noted that the terms "at least one," "one or more," or similar expressions indicating a feature or element may appear once or more, but are typically used only once when the corresponding feature or element is introduced. In most cases, the expressions "at least one" or "one or more" are not repeated when referring to the corresponding feature or element, but in fact, the corresponding feature or element may appear once or more.

[0181] Furthermore, as used herein, the terms “preferredly,” “more preferably,” “particularly,” “more particularly,” “specifically,” “more specifically,” or similar terms are used in combination with optional features without limiting the possibility of alternatives. Therefore, the features introduced by these terms are optional features and are not intended to limit the scope of the claims in any way. As those skilled in the art will recognize, the invention can be carried out by using alternative features. Similarly, features introduced by phrases such as “in embodiments of the invention” are intended to be optional features and do not limit any alternative embodiments of the invention, the scope of the invention, or the possibility of combining features introduced in this way with other optional or non-optional features of the invention.

[0182] In summary, and without excluding other possible embodiments, the following embodiments are conceivable:

[0183] Example 1. A method for determining whether an object corresponds to a living organism, the method comprising:

[0184] i) Illuminate the object with a beam pattern generated by at least one emitting device of the time-of-flight system, wherein the emitting device includes at least one array of light emitters;

[0185] ii) By using at least one detector of the time-of-flight system, at least one pattern image of the object is generated while the object is illuminated by the beam pattern;

[0186] iii) Extract live data from the pattern image using beam profile analysis, and determine whether the object corresponds to a living organism based on the live data.

[0187] Example 2. The method according to the previous example, wherein extracting live data includes extracting material data and / or extracting blood perfusion data and / or extracting at least one surface roughness measure.

[0188] Example 3. The method according to the previous embodiment, wherein extracting live data includes providing the pattern image to the model and / or receiving material data and / or blood perfusion data or surface roughness metrics from the model, wherein the model is a data-driven model, and wherein the model is configured, particularly parameterized and / or trained, based on historical pattern images and corresponding live data.

[0189] Example 4. The method according to any one of the foregoing two examples, wherein the extraction of material data includes generating one or more partial images from the pattern image.

[0190] Example 5. The method according to any one of the preceding three embodiments, wherein the extraction of material data includes generating a representation associated with a pattern image of the object, such as at least one tensor representing the pattern image or a dimensionality-reduced representation of the pattern image.

[0191] Example 6. The method according to any one of the foregoing embodiments, wherein the light emitters include one or more of at least one laser source, at least one light-emitting diode, or at least one laser diode.

[0192] Example 7. The method according to any one of the foregoing embodiments, wherein the transmitting device includes at least one imaging and / or collimating optics, the at least one imaging and / or collimating optics including at least one optical element selected from the group consisting of: at least one refractive lens; at least one metasurface lens; and at least one diffractive optical element (DOE).

[0193] Example 8. The method according to any one of the foregoing embodiments, wherein the detector includes at least one pixelated imaging element, the at least one pixelated imaging element including a plurality of optical sensors, wherein the optical sensors are arranged in a two-dimensional matrix.

[0194] Example 9. The method according to the previous embodiment, wherein the optical sensors are configured to measure the intensity of light from the object under beam pattern illumination.

[0195] Example 10. A system for determining whether an object corresponds to a living organism, wherein the system includes

[0196] - At least one time-of-flight system, comprising at least one transmitting device for generating a light beam pattern to illuminate the object, wherein the transmitting device includes at least one light emitter array, and wherein the time-of-flight system further includes at least one detector configured to generate at least one patterned image of the object while the object is illuminated by the light beam pattern.

[0197] - At least one processor configured to: extract live data from the pattern image using beam profiling analysis, and determine, based on the live data, whether the object corresponds to a living organism.

[0198] Example 11. The system according to the previous embodiment, wherein the system is configured to perform a method for determining whether an object corresponds to a living organism according to any of the foregoing embodiments relating to the method.

[0199] Example 12. A computer program including instructions that, when executed by a system according to any one of the foregoing embodiments relating to a system, cause the system to perform a method according to any one of the foregoing embodiments relating to a method.

[0200] Example 13. A computer-readable storage medium comprising instructions that, when executed by a system according to any one of the foregoing embodiments relating to a system, cause the system to perform a method according to any one of the foregoing embodiments relating to a method.

[0201] Example 14. A non-transient computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform the method according to any one of the foregoing embodiments relating to the method.

[0202] Example 15. A method implemented by a computer for authenticating a device, the method comprising: receiving a request for accessing at least one resource associated with the device and performing at least one authentication process, wherein the authentication process includes...

[0203] a. Determine the user's distance information by using a time-of-flight system, and determine whether the user is within or outside the scope of the authentication process by comparing the distance information with the working area;

[0204] b. If it is determined that the user is within the scope of the work, allow the user to access the resource; otherwise, if it is determined that the user is outside the scope of the work, deny the user access to the resource.

[0205] Example 16. The method according to the previous embodiment, wherein the authentication process further includes: determining whether the object corresponds to a living organism by executing the method for determining whether an object corresponds to a living organism according to any one of the foregoing embodiments relating to the method for determining whether an object corresponds to a living organism, wherein the authentication process includes: allowing the user to access the resource if it is determined that the user's pattern image corresponds to a living organism, otherwise denying the user access to the resource if it is determined that the user's pattern image does not correspond to a living organism.

[0206] Example 17. The method according to any one of the preceding two embodiments, wherein the authentication process further includes generating at least one floodlight image showing the associated user while the user is being illuminated by floodlight, and determining whether the user's identity corresponds to a verified identity based on the floodlight image, wherein the authentication process includes: allowing the user to access the resource if the user's identity corresponds to a verified identity, otherwise denying the user access to the resource if the user's identity does not correspond to a verified identity.

[0207] Example 18. According to the method described in the previous embodiment, in order to determine whether the user's identity corresponds to a verified identity based on the floodlight image, the similarity between at least one image feature vector obtained from the floodlight image and at least one template feature vector is considered, wherein the template vector is obtained from a template image, wherein the template image is generated during the registration process.

[0208] Example 19. The method according to any one of the foregoing four embodiments, wherein the device is selected from the group consisting of: television equipment; game consoles; personal computers; mobile devices, particularly mobile phones, and / or smartphones, and / or tablet computers, and / or laptop computers, and / or tablet computers, and / or virtual reality devices, and / or wearable devices such as smartwatches; or other types of portable computers.

[0209] Example 20. A system for authenticating users, including

[0210] - At least one time-of-flight system;

[0211] -At least one processor; and

[0212] - At least one memory configured to store instructions that, when executed by the processor, cause the system to perform the method for authenticating a user of a device according to any one of the foregoing embodiments relating to the method for authenticating a user of a device.

[0213] Example 21. The system according to the previous embodiment, wherein the system is further configured to determine whether an object corresponds to a living organism, wherein the time-of-flight system includes at least one emitting device for generating a beam pattern for illuminating the object, wherein the emitting device includes at least one light emitter array, wherein the time-of-flight system further includes at least one detector configured to generate at least one pattern image of the object while the object is illuminated by the beam pattern, wherein the processor is configured to: extract liveness data from the pattern image by using beam profiling analysis, and determine whether the object corresponds to a living organism based on the liveness data.

[0214] Example 22. The system according to the previous embodiment, wherein the system is configured to perform the method for determining whether an object corresponds to a living organism according to any one of the foregoing embodiments relating to the method for determining whether an object corresponds to a living organism.

[0215] Example 23. A computer program comprising instructions that, when executed by a system for authentication according to any one of the foregoing embodiments relating to a system for authentication, cause the system for authentication to perform a method for authenticating a user according to any one of the foregoing embodiments relating to a method for authenticating a user.

[0216] Example 24. A computer-readable storage medium comprising instructions that, when executed by a system for authentication according to any one of the foregoing embodiments relating to a system for authentication, cause the system for authentication to perform a method for authenticating a user according to any one of the foregoing embodiments relating to a method for authenticating a user.

[0217] Example 25. A non-transient computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform the method for authenticating a user according to any one of the foregoing embodiments relating to a method for authenticating a user.

[0218] Example 26. The use of a time-of-flight system for authenticating users and / or determining whether an object corresponds to a living organism.

[0219] Example 27. Images generated by the detectors of the time-of-flight system are used to authenticate users and / or determine whether an object corresponds to a living organism. Attached Figure Description

[0220] Further optional features and embodiments will be disclosed in more detail, preferably in conjunction with the dependent claims, in the following description of embodiments. As those skilled in the art will recognize, the respective optional features can be implemented independently and in any feasible combination. The scope of the invention is not limited to the preferred embodiments. Embodiments are schematically depicted in the accompanying drawings. The same reference numerals in these drawings refer to the same or functionally equivalent elements.

[0221] In the attached diagram:

[0222] Figure 1 Exemplary setups are shown for a system for determining whether an object corresponds to a living organism and for authenticating a user;

[0223] Figure 2 A flowchart illustrating an embodiment of a method implemented by a user's computer for authenticating a device; and

[0224] Figure 3 An example is shown of extracting liveness data from a patterned image using beam profiling analysis, and determining whether an object corresponds to a living organism based on the liveness data. Detailed Implementation

[0225] Figure 1 Exemplary setups are shown for a system 110 for determining whether an object corresponds to a living organism and a system 112 for authenticating a user.

[0226] Determining whether a user corresponds to a living organism can include distinguishing non-living organisms, such as distinguishing skin from non-skin materials (like silicon).

[0227] System 110 includes

[0228] - At least one time-of-flight system 114, the at least one time-of-flight system including at least one emitting device 116 for generating a light beam pattern to illuminate the object, wherein the emitting device 116 includes at least one light emitter array 118, wherein the time-of-flight system 114 further includes at least one detector 120, the at least one detector being configured to generate at least one pattern image 122 of the object while the object is illuminated by the light beam pattern (in Figure 3 (Example shown in the text)

[0229] - At least one processor 124, the at least one processor being configured to: extract live data from the pattern image by using beam profiling analysis, and determine, based on the live data, whether the object corresponds to a live organism.

[0230] System 112 for users of authentication devices includes

[0231] -At least one time-of-flight system 114;

[0232] - At least one processor 124; and

[0233] - At least one memory 126, configured to store instructions that, when executed by the processor 124, cause the system 112 to perform, for example, [actions]. Figure 2 The method for authenticating a user's device according to the present invention is described herein.

[0234] The device can be selected from the group consisting of: television equipment; game consoles; personal computers; mobile devices, especially mobile phones, and / or smartphones, and / or tablet computers, and / or laptop computers, and / or tablet computers, and / or virtual reality devices, and / or wearable devices such as smartwatches; or other types of portable computers.

[0235] Systems 110 and / or 112 used for authenticating users may be included in the user's device, for example, as a component of the user's device or as the user's device itself.

[0236] The following section combines the use of authentication, such as Figure 2 The components of systems 110 and 112 are described in the description of an embodiment of a method implemented by a user's computer of the device shown.

[0237] like Figure 2 As shown, the method includes receiving a request for accessing at least one resource associated with the device and performing at least one authentication process.

[0238] The certification process includes

[0239] - (Figure 128) The distance information of the user is determined by using a time-of-flight system, and the user is determined to be within or outside the scope of the authentication process by comparing the distance information with the scope of the work.

[0240] - (Figure 130) If it is determined that the user is within the scope of the work, the user is allowed to access the resource; otherwise, if it is determined that the user is outside the scope of the work, the user is denied access to the resource.

[0241] Access may include entering and / or using one or more functions associated with the device. Functions can be any function, such as accessing at least one element and / or at least one resource associated with the device or the device. Functions requiring user authentication may be predefined. One or more functions associated with the device may include unlocking the device and / or accessing an application preferably associated with the device and / or accessing a portion of an application preferably associated with the device. For example, the function may include accessing content on the device, such as content stored in the device's database and / or content that can be retrieved by the device. In embodiments, allowing a user to access a resource may include allowing a user to perform at least one operation with the device and / or system. Resources may be devices, systems, device functions, system functions, and / or entities. Additionally and / or alternatively, allowing a user to access a resource may include allowing a user to access entities. Entities may be physical entities and / or virtual entities. Virtual entities may be, for example, databases. Physical entities may be access-restricted areas. Access-restricted areas may be one of the following: secure areas, rooms, apartments, vehicles, portions of the examples mentioned above, etc. The device may be locked and can only be unlocked by an authorized user.

[0242] The device may include at least one communication interface, such as a user interface, configured to receive requests for access to at least one resource associated with the device, particularly for performing at least one operation requiring authentication on the device. The request for access may include at least one action and / or instance requesting access. Receiving a request may include, for example, a process of obtaining the request from a data source and / or the user interface. Receiving may be fully or partially automated. Receiving a request for access to one or more functions associated with the device can be performed using at least one communication interface. Receiving may include, for example, receiving at least one user input via at least one user interface (e.g., the device's display), and / or receiving requests from remote devices and / or the cloud (e.g., via device communication, such as via the internet). For example, the request may be generated or triggered by at least one user input (e.g., via entering a security number or other user unlocking action), and / or may be sent from remote devices and / or the cloud (e.g., via a connected account).

[0243] The authentication process can be executed using at least one authentication unit (e.g., including processor 124) configured to perform at least one authentication procedure for a user. Execution of the authentication process can be triggered and / or started by receiving a request.

[0244] The authentication process may include multiple steps. For example, the authentication process may include performing at least one face detection step. The face detection step may include analyzing at least one image of the user, which is generated, for example, by the detector 120 of the TOF system 114 or another camera.

[0245] The authentication process may further include generating at least one floodlight image showing the associated user while the user is being illuminated by the floodlight, and determining, based on the floodlight image, whether the user's identity corresponds to a verified identity. The authentication process may include: allowing the user to access the resource if the user's identity corresponds to a verified identity, and otherwise denying the user access to the resource if the user's identity does not correspond to a verified identity.

[0246] The method may include:

[0247] -Utilize floodlight illumination on users by using at least one floodlight source;

[0248] - Capture at least one floodlight image by using at least one image generation unit (such as detector 120 of TOF system 114 or another camera).

[0249] A floodlight image may include an image showing a user, particularly the user's face, while the user is being illuminated by the floodlight. The floodlight image can be generated by imaging and / or recording light reflected from objects and / or the user illuminated by the floodlight. The floodlight image showing the user may include at least a portion of the floodlight on at least a part of the user. For example, the illumination and imaging of the floodlight source may be synchronized, for example, by using at least one control unit.

[0250] The face detection step may include analyzing a floodlight image. For example, an authentication process may include performing at least one face detection using a floodlight image. Face detection can be performed locally on the device. However, face recognition (i.e., assigning an identity to a detected face) can be performed remotely, for example, in the cloud, especially when identification rather than just verification is required. User templates can be stored at a remote device, such as in the cloud, and do not need to be stored locally. This can be advantageous from a storage and security perspective.

[0251] The authentication process may include identifying a user based on a floodlight image. Identification may include assigning an identity to a detected face and / or at least one identity check and / or verification of the user's identity. Therefore, in particular, the authentication unit may forward data to a remote device. Alternatively or additionally, the authentication unit may perform user identification based on a floodlight image, particularly by running an appropriate computer program with corresponding functionality. Identification may include assigning an identity to a detected face and / or verifying the user's identity. Identification may include performing facial verification on the imaged face to confirm whether it is the user's face. Identifying the user may include matching the floodlight image (e.g., showing the outline of parts of the user, particularly parts of the user's face) with a template. To match the floodlight image with the template, the similarity between at least one image feature vector obtained from the floodlight image and at least one template feature vector may be considered and / or evaluated. This template vector may be obtained from a template image.

[0252] The template image can be generated during the registration process. The registration process may include at least one step of registering, particularly registering with the service. During registration, the template image can be generated under secure conditions to ensure that the generated template image is displayed to the user. The registration process may include at least one of the following steps: capturing the template image; recording personal data, etc.

[0253] Face detection may include analyzing floodlight images. Specifically, the analysis of floodlight images may include using at least one image recognition technique, particularly face recognition technology. Image recognition technology includes at least one process for identifying a user in an image. Image recognition may include using at least one technique selected from the following: color-based image recognition, for example using features such as template matching; segmentation and / or connected component (blob) analysis, for example using size or shape; machine learning and / or deep learning, for example using at least one convolutional neural network. Refer to the above description regarding image recognition.

[0254] For example, authentication may include identifying a user. Identification may include assigning an identity to a detected face and / or at least one identity check and / or verification of the user's identity. Identification may include performing facial verification on the imaged face to confirm whether it is the user's face. Identifying a user may include matching a flood image (e.g., showing the outlines of various parts of the user, particularly the outlines of various parts of the user's face) with a template (e.g., a template image generated during registration). Identifying a user may include determining whether the imaged face is the user's face, and in particular determining whether the imaged face corresponds to at least one image of the user's face stored in at least one memory of a device, for example. If the flood image can match the image template, authentication may be successful. If the flood image cannot match the image template, authentication may be unsuccessful.

[0255] As described above, the method proposes to consider distance information obtained and / or provided by the TOF system 114. In step a.128, the method may include determining the distance of an object to the detector 120 of the TOF system 114 by using the TOF system 114. The method may include determining whether the distance of the object to the detector 120 is within the working range, and allowing the user access to resources in response to determining that the distance is within the working range. The present invention allows for the consideration of distance information obtained by the TOF system for the authentication process. The detector 120 may be adapted to determine the distance and generate an image for authentication. Therefore, advantageously, space in the device can be saved by using the TOF system 114 for authentication.

[0256] Distance is also important for facial authentication because the facial recognition model is associated with a working range. The working range can specify the distance range between the user's face and detector 120 for generating an image of the user, on which the facial recognition model operates and / or is trained. In embodiments, the working range can specify at least one upper boundary and / or at least one lower boundary of the distance of the object from the detector and / or illumination source. The working range can be associated with the authentication process. The working range can include at least one value. This value can be numerical, particularly a positive value. An indication of the working range can be received, particularly before determining whether the distance is within or outside the working range of the authentication process. The indication of the working range can be adapted to determine whether the distance is within or outside the working range of the authentication process. The indication of the working range can be adapted to compare the distance with the working range.

[0257] Distance information can be determined directly in detector 120 of the TOF system 114. Determining distance information may include considering, for example, calibration data stored in memory 126 of detector 120. Detector 120 may include at least one communication interface configured to provide the user's image to another unit for further analysis and / or provide distance information to an authentication unit to check whether the distance is within or outside the operating range.

[0258] The authentication process may further include determining whether an object corresponds to a living organism by performing the method according to the invention for determining whether an object corresponds to a living organism.

[0259] Methods for determining whether an object corresponds to a living organism include:

[0260] i) (reference numeral 132) Illuminate the object using a beam pattern generated by a transmitting device 116 of the time-of-flight system 114, wherein the transmitting device 116 includes a light emitter array 118;

[0261] ii) (Ref. 134) At least one pattern image 122 of the object is generated while the object is being illuminated by the beam pattern using the detector 120 of the time-of-flight system 114;

[0262] iii) (Figure reference 136) Live data is extracted from the pattern image 122 by using beam profile analysis, and the live data is used to determine whether the object corresponds to a living organism.

[0263] The transmitters can be configured to emit light pulses. Each light transmitter can be configured to emit beams one after another (e.g., within time intervals), for example controlled by at least one control unit of the TOF system 114.

[0264] The transmitter may include one or more of at least one laser source, at least one light-emitting diode, or at least one laser diode. For example, the laser source may be at least one vertical-cavity surface-emitting laser (VCSEL).

[0265] For example, the transmitting device 116 may be or may include at least one radio frequency (RF) modulated light source. In this case, the light transmitter may be and / or may include at least one light-emitting diode or at least one laser diode. The transmitting device 116 may be configured to modulate light generated by the light transmitter using an RF carrier. The transmitting device may be configured to modulate light having an RF carrier frequency from 100 kHz to 300 MHz, preferably from 80 MHz to 200 MHz. For example, the transmitting device 116 may be configured to modulate light at a high speed of up to 100 MHz.

[0266] For example, the TOF system 114 may include at least one direct TOF imager. In this case, the light emitter may be and / or may include at least one laser source, such as at least one infrared laser source. This allows the illumination to be imperceptible. A single pulse per frame can be used, ranging from 10 to 100 Hz, preferably from 15 to 80 Hz, more preferably from 20 to 60 Hz. For example, a single pulse per frame, such as 30 Hz, can be used.

[0267] The transmitting device 116 may include at least one imaging and / or collimating optics 138, which includes at least one optical element selected from the group consisting of: at least one refractive lens; at least one metasurface lens; and at least one diffractive optical element (DOE). The optical element is configured to collimate and / or reproduce light.

[0268] The light beam illuminating the object can be reflected by the object and imaged by the detector 120 of the TOF system 114. The detector 120 can be and / or may include at least one image sensor. The detector 120 includes at least one pixelated imaging element, which comprises a plurality of optical sensors. The optical sensors can be arranged in a two-dimensional matrix. For example, the detector 120 may include at least one CMOS sensor or at least one CCD chip.

[0269] Each pixel of detector 120 can be configured to measure the time it takes for light to travel from the transmitter to the object and back to detector 120. For example, in the case of using an RF modulated light source, detector 120 of a TOF system can be configured to measure the phase shift of the carrier wave. For example, in the case of using the direct time-of-flight measurement principle, detector 120 can be configured to determine the time it takes for a single pulse to leave the transmitter and reflect back to detector. In particular, when operating in a so-called triggered mode, detector 120 can generate a 3D image that includes both spatial and temporal data. Detector 120 (e.g., per pixel) can be configured to measure the intensity of the incident light, particularly the amplitude of the modulated light on the detector.

[0270] Pattern image 122 may include an image showing at least a portion of the user's face, particularly within a corresponding region of interest included in the image, while the user is being illuminated by the pattern. Therefore, detector 120 can generate an intensity image of the object projected by the pattern. The pattern image can be generated by imaging and / or recording the light reflected from the object illuminated by the light pattern.

[0271] Therefore, this invention proposes using a TOF system 114 to generate images, particularly 2D images, such as grayscale images, for evaluation to extract live data via beam profiling analysis. The image can be processed in step iii), for example, by using beam profiling analysis, to obtain material information about the object.

[0272] Step iii) 136 includes extracting liveness data from the pattern image 122 using beam profiling analysis, and determining whether the object corresponds to a living organism based on the liveness data. Determining whether an object corresponds to a living organism based on the pattern image may include extracting liveness data from the pattern image. Specifically, this can be used to determine that the object is a human. In particular, extracting liveness data includes extracting material data and / or extracting blood perfusion data.

[0273] For example, determining whether an object corresponds to a living organism based on at least one pattern image may include extracting material data from the pattern image 122. Specifically, this may determine that the user is human. The material data may include information about the material type of the user detected in the pattern image 122. Extracting material data from the pattern image 122 may or may include generating material type and / or data derived from the material type. The material data may include information about the material type of the object detected in the pattern image 122.

[0274] Extracting material data from a pattern image may include beam profile analysis of the light spot. For information on beam profile analysis, see WO 2018 / 091649 A1, WO 2018 / 091638 A1 and WO 2018 / 091640 A1, WO 2020 / 187719 A1, WO 2023 / 156469 A1, and WO 2023 / 156315 A1, the entire contents of which are incorporated herein by reference. Determining whether an object corresponds to a living organism based on at least one pattern image may include determining whether the extracted material data corresponds to desired material data. Determining whether the material data corresponds to desired material data may include comparing the material data with the desired material data. In an example, skin may be compared as desired material data with non-skin materials or silicon as material data, and the result may be rejection because silicon or non-skin materials may differ from skin. Comparing the material data with the desired material data may include determining the similarity between the extracted material data and the desired material data. The desired material data may refer to predetermined material data. In the example, the desired material data could be skin. It can be determined whether the material data corresponds to the desired material data. In the example, the material data could be a non-skin material or silicon.

[0275] To extract material data, a complete pattern image 122 can be used. Alternatively, a partial image can be used. Material data extraction may include generating one or more partial images from the pattern image 122. For example, different regions of the pattern image can be selected as partial images. The partial images can be different from each other. In particular, the partial images can be non-overlapping. Using partial images allows for the collection of material data from different regions of the object (e.g., under different lighting conditions) and / or comparison of the extracted material data and / or generation of material maps and / or reduction of uncertainty in the obtained material data.

[0276] In an embodiment, extracting material data from a pattern image may include generating material type and / or data derived from the material type. Preferably, the extraction of material data may be based on the pattern image. Material data can be extracted using at least one model. Extracting material data may include providing the pattern image 122 to at least one model and / or receiving material data from the model. Refer to the above description regarding the extraction of material data.

[0277] Alternatively or concurrently, determining whether an object corresponds to a living organism based on at least one pattern image may include determining at least one blood perfusion measurement. In particular, this may determine that the human being is alive.

[0278] Determining the at least one blood perfusion measure may include determining at least one speckle contrast of the patterned image. Alternatively or additionally, determining the at least one blood perfusion measure may include determining the blood perfusion measure based on the determined at least one speckle contrast. As used herein, the term "speckle contrast" is a broad term and will be given its common and conventional meaning to those skilled in the art and is not limited to a specific or custom meaning. The term may specifically refer to, but is not limited to, the degree of variation in a speckle pattern generated by coherent light. The speckle pattern can be generated by an emitting device, particularly on an object. The speckle contrast can be a measure of the average contrast of the intensity distribution within a region of the speckle pattern. In particular, the speckle contrast K over a region of the speckle pattern can be expressed as the standard deviation σ versus the average speckle intensity. The ratio, that is,

[0279]

[0280] Speckle contrast can include speckle contrast values. These values ​​can range from 0 to 1. Blood perfusion can be determined based on speckle contrast. Blood perfusion can depend on the determined speckle contrast. If the speckle contrast changes, the blood perfusion obtained from it will also change accordingly. Blood perfusion can be a single number or value that represents the likelihood that the object is alive. To monitor changes in speckle contrast, multiple pattern images generated at different time points can be used.

[0281] To determine speckle contrast, a full-pattern image can be used. Alternatively, a portion of a pattern image can be used. Preferably, a portion of the pattern image represents a smaller area of ​​the pattern image than the full-pattern image.

[0282] In this embodiment, a data-driven model can be used to determine blood perfusion measurements. The data-driven model is parameterized and / or trained based on a training dataset. The training dataset may include pattern images and blood perfusion measurements. The data-driven model can be parameterized and / or trained based on the training dataset to output blood perfusion measurements based on received pattern images.

[0283] In an embodiment, determining whether an object corresponds to a living organism based on blood perfusion data may include determining whether a blood perfusion measurement corresponds to a human blood perfusion measurement. Determining whether a blood perfusion measurement corresponds to a human may include comparing the blood perfusion measurement to at least one predefined or predetermined range of blood perfusion measurement values, for example, stored in at least one database. If the extracted blood perfusion measurement falls within the redefined or predetermined range of blood perfusion measurement values, at least within tolerance, the object is determined to correspond to a living organism; otherwise, it does not.

[0284] Figure 3 From left to right, the diagram shows pattern image 122, the selected and extracted spot 140 of the pattern image for beam profiling analysis to extract live data, and (highly illustratively) an example of determining whether an object corresponds to a live organism based on live data 142.

[0285] The above method can be executed completely automatically, especially without user interaction.

[0286] List of reference numerals

[0287] 110 A system for determining whether an object corresponds to a living organism.

[0288] 112 System for Authenticating Users

[0289] 114 Flight Time System

[0290] 116 Launching Device

[0291] 118 optical emitter array

[0292] 120 detector

[0293] 122 pattern images

[0294] 124 processor

[0295] 126 memory

[0296] Step a of 128.

[0297] Step 130b.

[0298] 132 Step i)

[0299] Step 134 (ii)

[0300] Step iii) of 136

[0301] 138 Imaging and / or Collimating Optics

[0302] 140 Selected and extracted light spots

[0303] 142 Determine whether the object corresponds to a living organism.

Claims

1. A method for determining whether an object corresponds to a living organism, the method comprising: i) (132) Illuminating the object using a beam pattern generated by at least one transmitting device (116) of the time-of-flight system (114), wherein, The transmitting device (116) includes at least one optical emitter array (118). ii) (134) At least one pattern image (122) of the object is generated while the object is illuminated by the beam pattern using at least one detector (120) of the time-of-flight system (114). iii) (136) Extract live data from the pattern image (122) by using beam profile analysis, and determine whether the object corresponds to a live organism based on the live data.

2. The method according to the preceding claim, wherein, The live data indicates whether the object corresponds to a living organism, wherein extracting live data includes extracting material data and / or extracting blood perfusion data and / or extracting at least one surface roughness measure.

3. The method according to the preceding claim, wherein, Extracting live data includes providing the pattern image (122) to the model and / or receiving material data and / or blood perfusion data or the surface roughness metric from the model, wherein the model is a data-driven model, and wherein the model is configured, in particular parameterized and / or trained based on historical pattern images and corresponding live data.

4. The method according to any one of the preceding claims, wherein, These light emitters include one or more of at least one laser source, at least one light-emitting diode, or at least one laser diode.

5. The method according to any one of the preceding claims, wherein, The detector (120) of the time-of-flight system (114) is configured to detect incident light for time-of-flight analysis, wherein the detector (120) is configured to measure the time it takes for the light to travel from the emitting device (116) to the object and from the object to the detector (120), and to detect the light to generate the pattern image (122).

6. A system (110) for determining whether an object corresponds to a living organism, wherein, The system (110) includes - At least one time-of-flight system (114), the at least one time-of-flight system including at least one transmitting device (116) for generating a beam pattern to illuminate the object, wherein the transmitting device (116) includes at least one light emitter array (118), wherein the time-of-flight system (114) further includes at least one detector (120) configured to generate at least one pattern image (122) of the object while the object is illuminated by the beam pattern. - At least one processor (124) configured to: extract live data from the pattern image (122) by using beam profile analysis, and determine whether the object corresponds to a live organism based on the live data.

7. A method implemented by a user's computer for authenticating a device, the method comprising: Receives a request for accessing at least one resource associated with the device and performs at least one authentication process, wherein the authentication process includes a. (128) Determine the user's distance information by using the time-of-flight system (114), and determine whether the user is within or outside the scope of the authentication process by comparing the distance information with the scope of the work. b. (130) If it is determined that the user is within the scope of the work, allow the user to access the resource; otherwise, if it is determined that the user is outside the scope of the work, deny the user access to the resource.

8. The method according to the preceding claim, wherein, The authentication process further includes: determining whether the object corresponds to a living organism by performing the method for determining whether an object corresponds to a living organism according to any one of the foregoing claims relating to the method for determining whether an object corresponds to a living organism, wherein the authentication process includes: allowing the user to access the resource if it is determined that the user's pattern image (122) corresponds to a living organism, otherwise denying the user access to the resource if it is determined that the user's pattern image does not correspond to a living organism.

9. The method according to any one of the preceding two claims, wherein, The device is selected from the group consisting of: television equipment; game consoles; personal computers; mobile devices, especially mobile phones, and / or smartphones, and / or tablet computers, and / or laptop computers, and / or tablet computers, and / or virtual reality devices, and / or wearable devices such as smartwatches; or other types of portable computers.

10. A system (112) for authenticating users, comprising: - At least one time-of-flight system (114). - At least one processor (124); and - At least one memory (126) configured to store instructions that, when executed by the processor (124), cause the system (112) to perform the method for a user of an authentication device according to any one of the foregoing claims relating to a method for an authentication device.

11. The system (112) according to the preceding claim, wherein, The system (112) is further configured to determine whether an object corresponds to a living organism, wherein the time-of-flight system (114) includes at least one emitting device (116) for generating a beam pattern for illuminating the object, wherein the emitting device (116) includes at least one light emitter array (118), wherein the time-of-flight system (116) further includes at least one detector (120) configured to generate at least one pattern image of the object while the object is illuminated by the beam pattern, wherein the processor (124) is configured to: extract liveness data from the pattern image by using beam profile analysis, and determine whether the object corresponds to a living organism based on the liveness data.

12. A computer program comprising instructions, the instructions being: When the program is executed by the system (110) for determining whether an object corresponds to a living organism according to any of the preceding claims relating to a system for determining whether an object corresponds to a living organism, the system (110) is caused to execute the method for determining whether an object corresponds to a living organism according to any of the preceding claims relating to a method for determining whether an object corresponds to a living organism, and / or When the program is executed by the system (112) for authentication according to any one of the foregoing claims relating to the system for authentication, the system (112) for authentication performs the method for authenticating a user according to any one of the foregoing claims relating to the method for authenticating a user.

13. A computer-readable storage medium comprising instructions, the instructions being: When these instructions are executed by the system (110) for determining whether an object corresponds to a living organism according to any of the preceding claims relating to a system for determining whether an object corresponds to a living organism, the system (110) is caused to perform the method for determining whether an object corresponds to a living organism according to any of the preceding claims relating to a method for determining whether an object corresponds to a living organism, and / or When these instructions are executed by the system (112) for authentication according to any one of the preceding claims relating to the system for authentication, the system (112) for authentication is caused to perform the method for authenticating a user according to any one of the preceding claims relating to the method for authenticating a user.

14. A non-transient computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform the method for determining whether an object corresponds to a living organism as described in any one of the preceding claims, and / or When executed by one or more processors, the one or more processors are caused to perform the method for authenticating a user as described in any one of the preceding claims relating to the method for authenticating a user.

15. The time-of-flight system (114) is used for authenticating users and / or determining whether an object corresponds to a living organism.

16. The image generated by the detector (120) of the time-of-flight system (114) is used for authenticating the user and / or determining whether the object corresponds to a living organism.

Images