Biometric identification system, biometric identification method and electronic device

Abstract

The present disclosure aims to improve a biometric identification capability on the basis of a SWIR camera. A biometric identification system comprises a SWIR camera or is connected to a SWIR camera, and comprises an identification part. The SWIR camera collects a SWIR image on the basis of SWIR spectra. The identification part receives the SWIR image, identifies whether a biological part corresponding to the SWIR image is real skin or artificial skin at least on the basis of the brightness of the SWIR image, and outputs an identification result. A biometric identification method comprises: a step of collecting a SWIR image on the basis of SWIR spectra; and a step of receiving the SWIR image, identifying whether a biological part corresponding to the SWIR image is real skin or artificial skin at least on the basis of the brightness of the SWIR image, and outputting an identification result. An electronic device comprises: a memory, which stores computer-executable codes; and a processor, which executes the codes so as to implement the following operation: receiving a SWIR image collected on the basis of SWIR spectra; and identifying whether a biological part corresponding to the SWIR image is real skin or artificial skin at least on the basis of the brightness of the SWIR image, and outputting an identification result.

Description

Biometric identification system, biometric identification method and electronic device

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to and the benefits of Chinese Patent Application No. 202410268437.7 filed with the State Intellectual Property Office of the People's Republic of China on March 8, 2024, the entire contents of which are hereby incorporated by reference in their entirety. Technical Field

[0003] The present technology relates to a biometric authentication system, a biometric authentication method, and an electronic device, and more particularly to a biometric authentication system, a biometric authentication method, and an electronic device capable of improving biometric authentication capabilities based on a SWIR (short-wave infrared) camera. Background Art

[0004] Facial recognition has become extremely popular and can be applied in many fields. However, attempts to attack facial recognition have never ceased. For example, criminals sometimes wear human skin masks when committing crimes. Some human skin masks can highly simulate real human faces, even simulating blood vessels, pores, and skin textures such as wrinkles. These human skin masks can successfully defeat many conventional facial recognition systems (such as mobile phone lock screen systems, supermarket facial payment systems, company or community access control systems, bank ATMs or safe deposit boxes, etc.). These conventional facial recognition systems have difficulty distinguishing between real faces and human skin masks because they use RGB (red, green, and blue) cameras.

[0005] Similarly, in conventional biometric systems, biometric methods and electronic devices that identify fingerprints, palm prints, foot prints or irises, there is also the problem of difficulty in detecting (identifying) fake skin made of counterfeit materials such as fingerprint films, palm print films, foot print films or iris films. Summary of the Invention

[0006] [Technical problems to be solved]

[0007] Existing biometric recognition systems and methods capture images of biological parts, preprocess the images, extract features from the preprocessed images, match the extracted features with feature templates stored in a database, and then determine the identity of the individual in the image based on the similarity. However, existing biometric recognition systems and methods, using the RGB spectrum, capture very limited image information, making it difficult to distinguish between real skin and fake skin made from counterfeit materials (such as human skin masks, fingerprint films, palm print films, foot print films, or iris films).

[0008] Especially in low-light environments or in the presence of obstructions such as fog, smoke, haze, dust, plastic bags or glass, the images collected using the RGB spectrum are very unclear and the image details obtained are insufficient, making it more difficult to identify and recognize organisms.

[0009] Accordingly, the present disclosure intends to provide a biometric recognition system, a biometric recognition method, and an electronic device that can simply and efficiently distinguish between real skin and fake skin based on a SWIR camera, thereby improving biometric authentication capabilities.

[0010] [Technical solution to the problem]

[0011] To address the aforementioned issues, one aspect of the present disclosure provides a biometric recognition system comprising a SWIR camera or connected to a SWIR camera, and a discrimination unit. The SWIR camera captures SWIR images for each of at least one part of a living being based on the SWIR spectrum. The discrimination unit receives the SWIR images captured by the SWIR camera, discriminates whether the part corresponding to the SWIR image is real skin or fake skin based at least on the brightness of the SWIR image, and outputs the discrimination result.

[0012] Another aspect of the present disclosure provides an electronic device comprising a memory and a processor. The memory stores computer-executable code. The processor is configured to execute the code to perform the following operations: receiving a SWIR image, the SWIR image being acquired based on the SWIR spectrum for each of at least one part of a living being; and determining, based at least on the brightness of the SWIR image, whether the part corresponding to the SWIR image is real skin or fake skin, and outputting the determination result.

[0013] Another aspect of the present disclosure provides a biometric identification method, comprising a SWIR image acquisition step and a discrimination step. In the SWIR image acquisition step, a SWIR image is acquired for each of at least one part of a living being based on the SWIR spectrum. In the discrimination step, the acquired SWIR image is received, and based on at least the brightness of the SWIR image, the part corresponding to the SWIR image is discriminated as real skin or fake skin, and the discrimination result is output. [Beneficial Effects]

[0014] The biometric recognition system, biometric recognition method, and electronic device disclosed herein (which are sometimes collectively referred to as a SWIR recognition solution) for improving biometric authentication capabilities based on a SWIR camera have the following effects.

[0015] (1) The SWIR recognition scheme can simply and efficiently identify human skin masks and overcome the shortcomings of current ordinary face recognition systems.

[0016] (2) From the simulation results, the SWIR recognition scheme can identify human skin masks more easily than the recognition schemes using RGB cameras, near-infrared (NIR) cameras or depth cameras.

[0017] (3) Using a bandpass filter of 1400nm to 1500nm, the SWIR recognition scheme has the best effect in identifying human skin masks.

[0018] (4) Even for wet human skin masks, the SWIR recognition solution can identify them.

[0019] (5) Even when a person has makeup on his or her face, the SWIR recognition scheme can distinguish whether it is a real face or a human skin mask.

[0020] (6) Regarding the problem that the brightness of SWIR images may decrease as the distance to the face increases, the present disclosure also proposes the following three solutions: (i) using AE (Auto Exposure) on mode; (ii) using a light source array to obtain uniform light; and / or (iii) using depth information for joint judgment.

[0021] (7) The SWIR recognition scheme helps to improve the security level of the face recognition system to prevent attacks from human skin masks.

[0022] (8) SWIR recognition solutions can be applied to, for example, facial recognition payment systems, ATM (Automated Teller Machine) systems, and access monitoring systems for important places / ports (such as customs). BRIEF DESCRIPTION OF THE DRAWINGS

[0023] FIG1 is a block diagram showing a configuration example of a biometrics authentication system according to a first embodiment of the present disclosure.

[0024] FIG. 2 is a block diagram showing a configuration example of a SWIR camera in the biometric authentication system according to the first embodiment of the present disclosure.

[0025] FIG3 shows a comparison of images of real human faces and human skin masks using four cameras: an RGB camera, an NIR camera, a depth camera, and a SWIR camera.

[0026] FIG4 shows a comparison of images of a real human face and a human skin mask in the case of an RGB camera and in the case of a SWIR camera using different SWIR bandpass filters.

[0027] FIG. 5 is a graph showing quantitative evaluations in the case of an RGB camera and in the case of a SWIR camera using different SWIR bandpass filters.

[0028] FIG6 shows a comparison of images of a real human face, a dry human skin mask, and a wet human skin mask when a 1450 nm bandpass filter is used for the SWIR camera.

[0029] FIG7( a ) shows an image of a face with makeup captured using an RGB camera.

[0030] FIG7( b ), ( c ), and ( d ) show comparisons of images of a human face with makeup, a human face without makeup, and a human skin mask when a 1450 nm bandpass filter is used for the SWIR camera.

[0031] FIG8 is a block diagram showing a configuration example of a discrimination section in the biometrics authentication system according to the first embodiment of the present disclosure.

[0032] FIG9 is a block diagram showing a configuration example of a biometrics authentication system according to a second embodiment of the present disclosure.

[0033] FIG10 is a block diagram showing a configuration example of a biometrics authentication system according to a third embodiment of the present disclosure.

[0034] FIG. 11 is a block diagram showing a configuration example of an identification section in a biometrics identification system according to a third embodiment of the present disclosure.

[0035] FIG12 shows an example of three network structures constituting the MTCNN algorithm used when performing face alignment in the recognition unit.

[0036] FIG13 is a flowchart showing an operation of a face recognition system as an example of a biometric recognition system according to the third embodiment of the present disclosure.

[0037] FIG14 is a block diagram showing a schematic configuration of a biometrics authentication system according to a fourth embodiment of the present disclosure.

[0038] FIG15 is a flowchart showing an operation of a face recognition system as one example of a biometric recognition system according to a third modification of the fourth embodiment of the present disclosure.

[0039] FIG16 shows a light source array composed of a plurality of halogen lamps. DETAILED DESCRIPTION

[0040] Modes for carrying out the present technology (hereinafter, referred to as embodiments) will now be described. The description will be made in the following order.

[0041] 1. Biometric Authentication System of the First Embodiment

[0042] [Overview of a construction example of a biometric recognition system]

[0043] [light source]

[0044] [Example of SWIR camera configuration]

[0045] [The effect of SWIR camera]

[0046] [Three solutions to the problem of SWIR image brightness decreasing as the distance from the face increases]

[0047] [Example of the structure of the resolution unit]

[0048] 2. Biometric Authentication System of the Second Implementation

[0049] 3. Biometric Authentication System of the Third Implementation

[0050] [Configuration Example of Identification Unit]

[0051] 4. Biometric Authentication System of the Fourth Implementation

[0052] 5. Electronic devices

[0053] 6. Application Examples

[0054] The first embodiment according to the present disclosure will be described below.

[0055] [1. Biometric Authentication System of First Embodiment]

[0056] [Overview of a construction example of a biometric recognition system]

[0057] FIG1 is a block diagram illustrating an example configuration of a biometric recognition system according to a first embodiment of the present disclosure. As shown in FIG1 , biometric recognition system 100A primarily includes a SWIR camera 10 and a resolution unit 20. While FIG1 illustrates an example in which SWIR camera 10 is incorporated into biometric recognition system 100A, SWIR camera 10 may also be provided external to and connected to biometric recognition system 100A. This disclosure describes a case in which SWIR camera 10 is incorporated into biometric recognition system 100A.

[0058] A light source 2 disposed near a subject 1 emits illumination light toward the subject 1. A portion of the illumination light, after being reflected by the subject 1, enters the SWIR camera 10 as incident light. Although FIG1 illustrates the subject 1 as the upper body, the subject 1 may be at least one part of a living being, such as any combination of the face, fingers, palms, soles, and irises.

[0059] The SWIR camera 10 can capture spectral information in both the visible and SWIR bands of incident light. This disclosure aims to highlight the special effects of light in the SWIR band within the incident light. Therefore, the SWIR camera 10 can be equipped with an additional SWIR bandpass filter to transmit only light in the SWIR band (i.e., the SWIR spectrum). This allows the SWIR camera 10 to capture a SWIR image of the subject 1 based on the SWIR spectrum within the incident light. The details of the structure of the SWIR camera 10 will be further described later with reference to FIG2 .

[0060] The placement of the SWIR bandpass filter is not particularly limited, as long as it is located upstream of the light receiving surface (imaging surface) of the sensor portion of the SWIR camera 10 in the direction of light incidence (i.e., it allows only the SWIR spectrum of incident light to pass through, thereby forming an image on the light receiving surface). For example, the SWIR bandpass filter can be built into the SWIR camera 10, or alternatively, it can be located externally to the SWIR camera 10, between the lens of the SWIR camera 10 and the subject 1. By placing the SWIR bandpass filter externally to the SWIR camera 10, it can be easily maintained and replaced.

[0061] In the present disclosure, the spectral range of the SWIR bandpass filter may be, for example, 1350 nm to 1600 nm, preferably 1400 nm to 1600 nm, more preferably 1400 nm to 1550 nm, and most preferably 1400 nm to 1500 nm. The optimal spectral range of the SWIR bandpass filter may be approximately 1450 nm.

[0062] Under normal circumstances, the subject 1 imaged by the SWIR camera 10 should be real skin (e.g., a real face). However, criminals sometimes use fake skin made of counterfeit materials (e.g., masks). To detect fake skin and prevent crime, the biometric recognition system 100A pre-trains a classification model corresponding to each part of the biometric being recognized (e.g., face, fingers, palms, soles, irises, etc.) to serve as an image classifier capable of distinguishing between fake and real skin. In the present disclosure, the classification model can be an AI (artificial intelligence) classification model, a threshold classification model, or a traditional classification model such as an SVM (Support Vector Machine). In Figure 1, the classification unit 20 is equipped with a pre-trained classification model. After the SWIR image captured by the SWIR camera 10 is input to the classification unit 20, the classification unit 20 identifies the SWIR image based on the classification model and classifies it as real or fake skin. The structure of the classification unit 20 and the training of the classification model will be further explained later.

[0063] If the input SWIR image is classified as fake skin, the classification unit 20 outputs a classification result indicating “fake skin.” If the input SWIR image is classified as real skin, the classification unit 20 outputs a classification result indicating “real skin.”

[0064] [light source]

[0065] In the present disclosure, the SWIR camera may use, for example, a halogen lamp as a light source.

[0066] Halogen lamps emit light of a full wavelength range (similar to sunlight), and have the advantages of high light intensity and wide wavelength, and are therefore particularly suitable for analysis and detection in a wide wavelength range. In the present disclosure, the light source can be a single light source or multiple light sources.

[0067] Under the condition of a single artificial light source, the location of the light source can be determined based on a threshold function obtained as follows: first, the subject 1 is positioned 1 meter away from the light source 2 and the SWIR camera 10. Then, based on actual conditions, the optimal brightness threshold at 1 meter is determined as (the brightness of the real face image at 1 meter + the brightness of the mask image at 1 meter) / 2, and the threshold function is set as the optimal brightness threshold at 1 meter / the square of the distance. The distance here can be determined, for example, using depth information provided by a depth sensor.

[0068] When multiple light sources are used, their placement can be determined based on the application scenario and location characteristics. For example, the area where the light sources are located can be divided into several zones. A illuminance meter can be used to measure the illuminance in each zone. If the illuminance in each zone is similar, the light sources are considered to be optimally positioned.

[0069] [Example of SWIR camera configuration]

[0070] Generally speaking, visible light covers the wavelength range of 400nm to 780nm, while near-infrared (NIR) light covers the wavelength range of 780nm to 2526nm. Within NIR light, short-wavelength infrared (SWIR) light covers the wavelength range of 900nm to 1700nm. Therefore, SWIR cameras can capture images that conventional visible light sensors cannot.

[0071] FIG2 is a block diagram illustrating an example configuration of a SWIR camera in a biometric authentication system according to the first embodiment of the present disclosure. As shown in FIG2 , the SWIR camera 10 includes, for example, an optical element 101, an imaging element (sensor unit) 102, a signal processing unit 103, a display unit 105, and a storage unit 106. The optical element 101 includes a lens composed of one or more lenses, and guides imaging light (incident light) from a subject 1 toward the imaging element 102, forming an image on the light-receiving surface of the imaging element 102.

[0072] Although not shown, the imaging element 102 includes a semiconductor substrate in which a plurality of pixels are formed. Each pixel has a photoelectric conversion region. The photoelectric conversion region photoelectrically converts received incident light to generate a signal having a charge corresponding to the amount of received incident light as an (analog) pixel signal.

[0073] In the foregoing description, an example is described in which a separate SWIR bandpass filter is formed and then installed between the optical element 101 and the imaging element 102 within the SWIR camera 10, or is removably installed outside the SWIR camera 10 and positioned between the optical element 101 and the subject 1. As another alternative, instead of forming a separate SWIR bandpass filter, a SWIR filter layer can be laminated on the light-incident side of the semiconductor substrate of the imaging element 102 of the SWIR camera 10. The spectral range of this SWIR filter layer can be the same as that of the aforementioned SWIR bandpass filter.

[0074] The signal processing unit 103 performs various signal processing on the pixel signals output from the imaging element 102. The image (image data) obtained as a result of the signal processing performed by the signal processing unit 103 can be provided to the display unit 105 for display, or provided to the storage unit 106 for storage (recording). For example, the signal processing unit 103 may include an analog-to-digital (A / D) converter 131 and a digital signal processor (DSP) circuit 132. The A / D converter 131 converts the analog pixel signals from the imaging element 102 into digital pixel signals, and the DSP circuit 132 performs processing such as amplification, correction, and calculation on the digital pixel signals from the A / D converter 131, and then outputs the processed signals.

[0075] [The effect of SWIR camera]

[0076] To confirm the effectiveness of SWIR cameras, the inventors of this disclosure conducted various experimental tests. For ease of explanation, these experimental examples use facial recognition as an example of biometric identification. In this case, the fake skin made from counterfeit materials is referred to as a "human skin mask" or simply a "mask."

[0077] <1. First Experimental Example: Comparison of SWIR Cameras with RGB Cameras, NIR Cameras, and Depth Cameras>

[0078] The following illustrates the effectiveness of SWIR cameras in distinguishing between images of real faces and human skin masks by comparing them with RGB cameras, NIR cameras, and depth cameras.

[0079] Figure 3 shows a comparison of images of a real face and a human skin mask using four different cameras: an RGB camera, an NIR camera, a depth camera, and a SWIR camera. In Figure 3, the upper image corresponds to the image of a real face, and the lower image corresponds to the image of a human skin mask. Figure 3 (a) corresponds to the RGB camera, where the camera model used is the iPhone 14 pro max. Figure 3 (b) corresponds to the NIR camera, where the spectral range of the NIR bandpass filter used is approximately 940 nm. Figure 3 (c) corresponds to the depth camera. Figure 3 (d) corresponds to the SWIR camera, where the camera model used is the Omron STC-LBS132POE-SWIR, the sensor is the IMX990, the lens is the Kowa LM8HC-VIS-SW, and the spectral range of the SWIR bandpass filter is approximately 1450 nm. Both images in Figures 3 (b) and (c) are from indirect time-of-flight (iToF) imaging technology.

[0080] As can be seen in Figure 3, with the RGB camera, the image of the real face and the image of the human skin mask have similar brightness. The same is true for the NIR camera and the depth camera. Compared to the RGB camera, NIR camera, and depth camera, the SWIR camera can have a more obvious difference in the brightness of the images of the real face and the human skin mask. The reason is that the real face is rich in liquids such as water under the skin, and water absorbs light in the SWIR band (for example, 1450nm in this case), so the resulting SWIR image is almost black. In contrast, the counterfeit material used to make the mask has no water (or very little water content), so the resulting SWIR image is relatively bright. Using this principle, the SWIR camera can simply and efficiently distinguish between real skin and fake skin.

[0081] In Table 1 below, the present disclosure provides a quantitative evaluation of the image brightness of a real human face and a human skin mask and the difference between the two under four conditions: using an RGB camera, an NIR camera, a depth camera, and a SWIR camera.

[0082] [Table 1] Quantitative evaluation of RGB camera, NIR camera, depth camera and SWIR camera

[0083] In the above Table 1, the Y value represents the image brightness value of the object, and m (meter) is the unit of the depth information (distance) of the object.

[0084] It is also obvious from Table 1 that when using a SWIR camera to shoot, the brightness difference between the case with a human skin mask and the case without a human skin mask (real human face) is very obvious, so it is easy to detect whether a human skin mask is worn.

[0085] <2. Second Experimental Example: Using Different SWIR Bandpass Filters>

[0086] During the image acquisition process of the SWIR camera, a SWIR bandpass filter can be used to perform imaging based on the SWIR spectrum. The inventors of the present disclosure tested the effects of different SWIR bandpass filters on the SWIR camera. Figure 4 shows a comparison of images of a real human face and a human skin mask in the case of an RGB camera and when the SWIR camera uses different SWIR bandpass filters. In Figure 4, the upper figure corresponds to the case of the image of a real human face, and the lower figure corresponds to the case of the image of a human skin mask. Specifically, (a1) in Figure 4 shows a comparison of the images of a real human face and a human skin mask captured when the camera model of the RGB camera is iPhone14 pro max. (a2) to (a7) in Figure 4 show a comparison of the images of a real human face and a human skin mask captured when the SWIR camera uses different SWIR bandpass filters. More specifically, (a2) to (a6) in FIG4 are cases where a SWIR bandpass filter is used (the spectral ranges are 1300 nm, 1350 nm, 1450 nm, 1550 nm, and 1600 nm, respectively), while (a7) in FIG4 is a case where a SWIR bandpass filter is not used (the spectral range is 300 nm to 1700 nm).

[0087] As shown in Figure 4, with an RGB camera, the image of a real face and the image of a human skin mask have similar brightness, making it difficult to distinguish the two based on image brightness. With a SWIR camera, the difference in brightness between the image of a real face and the image of a human skin mask is more pronounced, making it easier to distinguish between the two. Based on subjective judgment, the test results show that SWIR cameras have better resolution performance when the bandpass filter's spectral range is 1400nm to 1650nm. Furthermore, SWIR cameras have even better resolution performance when the bandpass filter's spectral range is approximately 1450nm, 1550nm, or 1600nm.

[0088] Corresponding to FIG4 , the present disclosure provides in Table 2 below a quantitative evaluation of the image brightness of a real human face and a human skin mask and the difference between the two in an RGB camera and a SWIR camera using different SWIR bandpass filters.

[0089] [Table 2] Quantitative evaluation of RGB cameras and SWIR cameras using different SWIR bandpass filters

[0090] Figure 5 is a graph showing the quantitative evaluation in the case of an RGB camera and in the case of a SWIR camera using different SWIR bandpass filters, where the vertical axis (percentage = (Y value of the human skin mask - Y value of the real face) / Y value of the human skin mask) corresponds to the difference (Gap) in Table 2.

[0091] As shown in Table 2 and the quantitative evaluation curves in Figure 5, when a 1450nm bandpass filter is used, the image brightness (Y value) of the human skin mask captured by the SWIR camera is the largest difference between that of the real human face (as indicated by the circle in Figure 5). Furthermore, this difference is also significant when using a 1550nm or 1600nm bandpass filter.

[0092] <3. Third Experimental Example: Wet Human Skin Mask>

[0093] The previous experimental example described the case where the human skin mask was free of moisture (or contained very little moisture). Due to concerns that criminals might use a wet human skin mask to attack the biometric recognition system, the inventors of the present disclosure also tested the detection performance of the SWIR camera on a wet human skin mask.

[0094] Figure 6 shows a comparison of images of a real human face, a dry human skin mask, and a wet human skin mask using a 1450nm bandpass filter on the SWIR camera. Table 3 below provides a quantitative assessment of the image brightness (Y value) corresponding to (a), (b), and (c) in Figure 6, respectively.

[0095] [Table 3] Quantitative evaluation of SWIR cameras using a 1450nm bandpass filter

[0096] The wet human-skin mask produced by the inventors of this disclosure contains a significant amount of moisture on its surface. As shown in Table 3, although the image brightness of this wet human-skin mask is approximately 10% lower than that of a dry human-skin mask, it is still significantly brighter than that of a real human face. This means that the SWIR camera of this disclosure can easily identify a wet human-skin mask.

[0097] <4. Fourth Experimental Example: Makeup on the Face>

[0098] Often, the person to be identified has makeup on their face. Concerned that the makeup might cover the real face and cause false detection by the SWIR camera, the inventors of the present disclosure also tested the detection effect of the SWIR camera for situations where makeup is on the face.

[0099] Figure 7(a) shows an image of a face with makeup, captured using an RGB camera. Concealer, foundation, blush, highlighter, liquid shadow, eye shadow, and lipstick are applied to different parts of the face. Figures 7(b), (c), and (d) compare images of a face with makeup, a face without makeup, and a human skin mask, captured using a SWIR camera with a 1450nm bandpass filter. Table 4 below provides a quantitative assessment of the image brightness (Y value) corresponding to Figures 7(b), (c), and (d).

[0100] [Table 4] Quantitative evaluation of SWIR cameras using a 1450nm bandpass filter

[0101] As shown in Table 4, while adding some makeup to a human face slightly increases the SWIR image brightness compared to a face without makeup, this brightness is still much lower than the image brightness of a human skin mask. This means that the SWIR camera of the present disclosure can easily distinguish between a real human face with makeup and a human skin mask.

[0102] [Three solutions to the problem of SWIR image brightness decreasing as the distance from the face increases]

[0103] In the application examples of this disclosure, regarding SWIR image brightness, if only a single artificial light source is used, the SWIR image brightness decreases as the distance to the subject (e.g., a face) increases. This complicates the setting of the brightness threshold used to distinguish between a human skin mask and a real face. To address this issue, this disclosure proposes the following three methods.

[0104] (1) When the SWIR camera is capturing images, the AE on mode is used and a fixed AE target value is set. This way, the brightness of the SWIR image does not change much as the distance from the face increases.

[0105] (2) When the SWIR camera acquires images, a light source array is used as the light source to generate uniform light in the application example.

[0106] In the application example disclosed in the present invention, the following light source array is preferably adopted: in the light source array, multiple (more than 2) light sources of the same type are used. The circle in Figure 16 shows a light source 2 composed of a single halogen lamp, and Figure 16 shows a light source array composed of multiple halogen lamps. The arrangement of the light source array is not particularly limited in the present disclosure, but can be determined according to the usage scenario and location characteristics of the application example. For example, the light sources used to constitute the light source array can be placed around the working area of ​​the application example, and the light source array can be arranged to produce uniform light for the working area. Using a illuminance meter to check whether the illumination in each place in the working area is similar can help confirm whether the light source array produces uniform light. In this environment, because of the presence of uniform light, the brightness of the facial area will not change no matter where the facial area is or how far it is from the SWIR camera.

[0107] This approach not only solves the problem in the application example disclosed herein, but also solves the problem in any other SWIR application example (such as food sorting) that uses a single artificial light source. The uniform light from the light source array can ensure that the SWIR image brightness remains constant in any application example.

[0108] (3) When distinguishing between real human faces and human skin masks based on SWIR image brightness, depth (distance) information (which can be obtained by, for example, a ToF (Time of Flight) camera or a dual camera) is used to estimate the SWIR image brightness at different distances.

[0109] The SWIR image brightness can be estimated using the formula Y1×(1 / L^2), where Y1 is the SWIR image brightness of a face 1 meter away using a single artificial light source, and L is the facial distance. Furthermore, the brightness threshold for distinguishing between human skin masks and real faces can be set as a function of facial distance.

[0110] First, this disclosure requires determining an appropriate brightness threshold T1 for the 1m distance application example. Because facial distance L (depth) can be obtained using a ToF camera or dual RGB cameras, this disclosure sets a threshold function T = T1 × (1 / L^2). This threshold function T is applicable to the application examples of this disclosure at any distance.

[0111] This approach not only solves the problems in this application example, but also any other SWIR application using a single artificial light source (such as food sorting). By combining depth information with the threshold function T disclosed in this disclosure, no side effects occur when the SWIR image brightness changes with different distances.

[0112] [Example of the structure of the resolution unit]

[0113] Figure 8 is a block diagram illustrating an example of the structure of a discrimination unit in a biometric authentication system according to the first embodiment of the present disclosure. As shown in Figure 8 , the discrimination unit 20 includes, for example, an ROI frame setting unit (face detection unit) 201, an ROI frame cropping unit 202, and a classification unit 203. This description will continue using face recognition as an example.

[0114] The ROI frame setting unit (face detection unit) 201 detects (e.g., manually) the size and position of a face in the SWIR image from the SWIR camera 10, thereby determining "where the face is." If a face is present in the SWIR image, the ROI frame setting unit (face detection unit) 201 sets a bounding box within the SWIR image that encompasses only the face (i.e., encircles the face with an ROI frame (regions of interest box)) and outputs the coordinates of the ROI frame. The SWIR image, including the ROI frame, is then output to the ROI frame cropping unit 202.

[0115] The ROI frame cropping unit 202 crops the ROI frame image containing only the face from the SWIR image to facilitate classification, conversion, feature extraction, analysis or recognition in subsequent stages.

[0116] The ROI frame image cropped by the ROI frame cropping unit 202 is input to the classification unit 203, which classifies the ROI frame image as real skin or fake skin based on a classification model (such as an AI classification model, a threshold classification model or a traditional classification model) and outputs the classification result.

[0117] It should be understood that the ROI frame setting unit (face detection unit) 201 and the ROI frame cropping unit 202 do not have to be set in the resolution unit 20, but can also be set in series between the SWIR camera 10 and the resolution unit 20, or can be merged into a preprocessing unit (not shown) located between the SWIR camera 10 and the resolution unit 20.

[0118] As a modified example, a conversion unit (not shown) may be provided between the ROI frame setting unit (face detection unit) 201 and the ROI frame cropping unit 202, or between the ROI frame cropping unit 202 and the classification unit 203. This conversion unit converts the SWIR image including the ROI frame from the ROI frame setting unit (face detection unit) 201 or the ROI frame image from the ROI frame cropping unit 202 into a recognizable image, so that feature extraction can be performed from the recognizable image, and the extracted features can be used together with image brightness for classification or recognition.

[0119] The following two examples will be described as examples of the classification model used in the classification unit 203 .

[0120] <1. Example of AI classification model>

[0121] The classification unit 203 shown in FIG8 may include an AI (artificial intelligence) classification model. In addition to classification based on the brightness of the SWIR image (i.e., the water content of the subject), the AI ​​classification model may also, if necessary, incorporate object features (e.g., facial contours, facial features, shape and position, skin texture, skin color, etc.) that can be extracted from the SWIR image. The following describes an example of the AI ​​classification model training process, which includes the following stages.

[0122] (1) Dataset preparation

[0123] (1-1) Data collection step: Collect sufficient data related to the AI ​​classification model to be trained from reliable data sources. For example, in the present disclosure, it is necessary to train an image classifier that can distinguish between masks and real faces, so a large amount of image data of masks and real faces must be collected.

[0124] (1-2) Data cleaning step: After collecting data, irrelevant data is removed and the accuracy and consistency of the data are ensured through deduplication, missing value processing, outlier filtering, data standardization, error correction, etc.

[0125] (1-3) Data division step: After cleaning the data, the data is divided into a training data set, a validation data set, and a test data set. The training data set is used to train the AI ​​classification model, the validation data set is used to verify the accuracy of the AI ​​classification model or optimize the AI ​​classification model, and the test data set is used to test the performance of the AI ​​classification model.

[0126] (2) Model design or selection stage: Design or select a machine learning algorithm suitable for the AI ​​classification model to be trained.

[0127] (2-1) Regarding image classification algorithms: In the present disclosure, it is necessary to distinguish between images of masks and real faces. Therefore, a CNN (Convolutional Neural Networks) architecture, which is one of the representative algorithms for deep learning, can be selected, such as the MobileNet architecture or the EfficientNet architecture.

[0128] (2-2) Regarding detection and feature extractor algorithms: For example, you can choose VGGNet (a deep convolutional neural network model proposed by the Computer Vision Geometry Group of Oxford University), ResNet (a deep neural network model based on residual learning proposed by Microsoft Research), or Transformer (a deep learning model based on the self-attention mechanism proposed by Google), etc.

[0129] (3) Model training phase: Use the training dataset and deep learning architecture to train the model.

[0130] (3-1) Find a suitable loss function for classification to distinguish masks from real faces.

[0131] For classification training, for example, a cross entropy loss function may be used.

[0132] Regarding the training of the detector and feature extractor, for example, an L1 loss function may be used.

[0133] (3-2) Find a suitable optimization function, such as the SGD (Stochastic Gradient Descent) algorithm or the Adam (Adaptive Moment Estimation) algorithm, to reduce the loss.

[0134] (3-3) Other parameters such as learning rate and batch size can be set to improve the training process.

[0135] (4) Model optimization and evaluation phase

[0136] (4-1) A validation dataset can be used to verify the accuracy of the model or to optimize the model.

[0137] (4-2) Techniques such as cross-validation and grid search can be used to adjust model parameters to improve model performance.

[0138] (4-3) The test dataset can be used to evaluate the performance of the model such as accuracy, precision, and recall.

[0139] (5) Model deployment phase: After completing the training, optimization, and evaluation of the model, the model is deployed to the application or device of the biometric system (in this example, the classifier 203).

[0140] When the ROI frame image cropped by the ROI frame cropping unit 202 is input into the classifier 203 , the classifier 203 can distinguish whether the face in the ROI frame image is a mask or a real face based on the deployed AI classification model.

[0141] <2. Example of Threshold Classification Model>

[0142] In the classification unit 203 shown in FIG8 , a threshold classification model or a traditional classification model such as an SVM (Support Vector Machine) can be deployed. For example, in the example where the classification unit 203 is deployed with a threshold classification model, a threshold for distinguishing whether the part corresponding to the input SWIR image is a mask or a real face can be pre-set based on the threshold classification model. The threshold can be an image brightness threshold. If the input SWIR image has a brightness value lower than the threshold, it is classified as a real face by the threshold classification model; if the input SWIR image has a brightness value above the threshold, it is classified as a mask by the threshold classification model. Of course, the threshold classification model trained for different parts of a living being (such as the face, fingers, palms, soles, irises, etc.) can set different corresponding thresholds for distinguishing whether the part belongs to real or fake skin.

[0143] As a variation, more than two brightness thresholds can be set. If the brightness of the input SWIR image is above the highest threshold, it is judged as fake skin and an alarm is output. If the brightness of the input SWIR image is below the lowest threshold, it is judged as real skin, and face recognition is directly performed and the recognition result is output. If the brightness of the input SWIR image is between the highest and lowest thresholds or within other intermediate threshold intervals, other features other than brightness (such as facial contours, facial features, texture, color, etc. extracted from the SWIR image) can be further combined for judgment.

[0144] When the ROI frame image cropped by the ROI frame cropping unit 202 is input into the classifier 203, the classifier 203 can distinguish whether the part corresponding to the image is a mask or a real face by comparing the brightness of the image with the threshold based on the deployed threshold classification model.

[0145] [2. Biometric Authentication System of Second Embodiment]

[0146] FIG9 is a block diagram showing an example of the construction of a biometric identification system according to a second embodiment of the present disclosure. The biometric identification system 100B according to the second embodiment differs from the biometric identification system 100A according to the first embodiment in that a prompt unit 50a is further provided. When the resolution result from the resolution unit 20 indicates that the area corresponding to the input SWIR image is fake skin, the prompt unit 50a provides a prompt by generating a specific sound, image, and / or text. For example, the prompt unit 50a may emit an alarm sound and / or may display an image or text message indicating "fake face" or "not allowed to pass."

[0147] [3. Biometric Authentication System of the Third Embodiment]

[0148] FIG10 is a block diagram showing a configuration example of a biometrics authentication system according to a third embodiment of the present disclosure.

[0149] As shown in FIG. 10 , the biometric authentication system 100C of the third embodiment differs from the biometric authentication system 100A of the first embodiment in that a biometric database storage unit 40 , a monitor 50 b , and an identification unit 60 are further provided.

[0150] In the third embodiment, if the input SWIR image is classified as fake skin, the classification unit 20 outputs the classification result indicating "fake skin" to the monitor 50b. All or part of the functions of the display unit 50a described in the second embodiment can be incorporated into the monitor 50b in the third embodiment. If the input SWIR image is classified as real skin, the classification unit 20 outputs the classification result indicating "real skin" and the SWIR image of the real skin (here, the ROI frame image) to the recognition unit 60.

[0151] When the recognition unit 60 receives the SWIR image output from the resolution unit 20, it identifies the identity of the living being in the SWIR image based on the pre-stored biometric image data template obtained from the biometric database storage unit 40 and outputs the identification result to the monitor 50b for display. Additionally or alternatively, the monitor 50b may emit a "pass" prompt tone, display an image or text message indicating "pass" or "this is the person," or provide instructions for unlocking the door.

[0152] In the biological database storage unit 40 , an image of at least one part of at least one biological user is stored in advance in a one-to-one correspondence with the identity (eg, name) of the user.

[0153] [Configuration Example of Identification Unit]

[0154] Figure 11 is a block diagram illustrating an example configuration of a recognition unit in a biometric recognition system according to a third embodiment of the present disclosure. As shown in Figure 11 , recognition unit 60 includes, for example, a preprocessing unit 601, an alignment unit 602, a representation unit 603, and a matching unit 604. The following description continues using face recognition as an example.

[0155] The SWIR image of the area identified as real skin from the identification unit 20 (as mentioned above, this is an ROI frame image containing only a human face) is input to the preprocessing unit 601. The preprocessing unit 601 performs preprocessing such as noise reduction on the ROI frame image, and outputs the ROI frame image obtained after preprocessing to the alignment unit 602.

[0156] If the details of the information in the ROI frame image are not clear enough and therefore difficult to identify, then preferably, the pre-processing unit 601 can also perform an image enhancement function. For example, the image enhancement function can enhance image details through methods such as sharpening, high-pass filtering, increasing image brightness, or increasing image contrast. As an example, the image enhancement function can be incorporated into the pre-processing unit 601. As another example, the image enhancement function can also be implemented by an image enhancement unit (not shown) arranged between the resolution unit 20 and the pre-processing unit 601 of the recognition unit 60, or between the pre-processing unit 601 and the alignment unit 602.

[0157] The face of the same biological subject may show different postures and expressions in different image sequences, which is not conducive to face recognition. Therefore, an alignment unit 602 is provided to perform face alignment, so that the ROI frame images received from the preprocessing unit 601 are transformed into a unified, standardized angle and posture. Regarding the face alignment algorithm, the examples provided here can be, for example, the MTCNN (Multi-task convolutional neural network) algorithm or the LAB (Look at Boundary) algorithm. Figure 12 shows an example of three network structures used to constitute the MTCNN algorithm used when performing face alignment in the recognition unit. The alignment unit 602 can, for example, use the MTCNN algorithm to perform face alignment. Among the three network structures used to constitute the MTCNN algorithm, P-Net is a candidate network (proposal network), R-Net is a refinement network (refine network), and O-Net is an output network (output network).

[0158] In the alignment unit 602, the face alignment process, image conversion process, feature point selection and positioning process, and cropping process can be performed sequentially, or the image conversion process, feature point selection and positioning process, feature point-based face alignment process, and cropping process can be performed sequentially, but the present disclosure is not limited to these two sequences. For example, if the image conversion process and feature point selection and positioning process have already been completed in the resolution unit 20, there is no need to repeat these processes in the alignment unit 602. Only the former of these two sequences will be used as an example for detailed description. During the face alignment process, for the ROI frame image from the preprocessing unit 601, the alignment unit 602 transforms the face position in the ROI frame image through algorithms such as affine transformation or perspective transformation to align the face; during the image conversion process, the alignment unit 602 converts the ROI frame image after face alignment into a recognizable image format; during the feature point selection and positioning process, the alignment unit 602 selects facial feature points (such as eyebrows, eyes, nose, mouth and facial contours, etc.) from the recognizable ROI frame image by using grayscale image processing technology and marks the position coordinates of these key positioning points; in addition, during the cropping process, the alignment unit 602 further crops the ROI frame image into a face image that only retains the key positioning points.

[0159] Continuing with reference to FIG11 , the standardized facial image from the alignment unit 602 is input to the representation unit 603 . The representation unit 603 converts the pixel values ​​of the facial image into a compact, discriminable feature vector through feature modeling (e.g., encoding). Ideally, the faces of the same biological subject should be mapped to similar feature vectors in all cases.

[0160] After batch preprocessing, large amounts of image data from the biometric database storage unit 40 are subjected to batch feature extraction to obtain image data templates. Table 5 below shows an example of an image data template in which the extracted facial feature vectors are stored in a one-to-one correspondence with biometric identities (e.g., names) in the facial database. However, the image data templates used in this disclosure are not limited to this example.

[0161] [Table 5] An example of an image data template

[0162] Next, the matching unit 604 compares and matches the discriminable feature vector from the representation unit 603 with the image data template obtained from the biological database storage unit 40, determines whether the two belong to the same biological subject based on the gap or similarity score, and outputs the recognition result to the monitor 50b for display.

[0163] As some variations, the pre-processing unit 601 and the alignment unit 602 in the third embodiment do not have to be provided in the recognition unit 60, but may be provided upstream of the recognition unit 60, or may be omitted depending on the needs of actual application examples.

[0164] Figure 13 shows an operational flow chart of a facial recognition system, serving as an example of a biometric recognition system according to the third embodiment of the present disclosure. For example, this operational flow chart includes steps (b1) through (b11) as described below. In steps (b1) through (b3) of Figure 13 , the upper diagram corresponds to a real human face, while the lower diagram corresponds to a human skin mask. Details of steps (b1) through (b11) are described below.

[0165] (b1) Image Input Step: In this step, the user (the person to be recognized, such as a human) places their upper body close to the SWIR camera lens, positions their face within the lens's viewfinder, and maintains a direct gaze. Once the blue progress bar completes and displays "All Ready," a SWIR image of the user is acquired. The imaging light used here is the SWIR spectrum of the incident light.

[0166] (b2) Performing face detection to set the ROI frame: Performing detection (e.g., manual detection) on the SWIR image obtained in step (b1). If a face is found, manually setting the ROI frame to include the face, and outputting the user's SWIR image including the ROI frame.

[0167] (b3) performing a cropping step: cropping the SWIR image including the ROI frame obtained in step (b2) into an ROI frame image containing only the face.

[0168] (b4) The step of performing classification using a classification model: that is, identifying or judging the ROI frame image obtained in step (b3) to determine whether it is a real face or a mask.

[0169] If the image is identified as a real face in step (b4), the recognition continues with subsequent steps (b5) to (b11). Conversely, if the image is identified as a mask in step (b4), the facial recognition system sounds an alarm and outputs an image or text message indicating a "fake face," without further processing. In other words, in the face recognition system of the present disclosure, if the input image is identified as a mask through the classification model in step (b4), the image will not be accepted.

[0170] In the classification step of step (b4), classification can be performed using an AI classification model such as MobileNet-v2, EfficientNet, or EfficientNet-Lite, or a threshold classification model can be used for classification, or a traditional classification model such as SVM (Support Vector Machine) can be used for classification.

[0171] (b5) Image enhancement step: that is, for the ROI frame image identified as a real face in step (b4), the details of the ROI frame image are enhanced by methods such as sharpening, high-pass filtering, increasing image brightness or increasing image contrast.

[0172] (b6) Perform a transformation step: convert the enhanced ROI frame image in step (b5) into a recognizable image format, select facial feature points in the recognizable image (key positioning points such as eyebrows, eyes, nose, mouth and facial contours) and mark the position coordinates of these key positioning points.

[0173] (b7) further performing a cropping step: further cropping the ROI frame image obtained in step (b6) into a face image that only retains the key facial positioning points.

[0174] (b8) Performing the step of aligning facial features: that is, for the facial image obtained in step (b7), correcting the posture of the facial image, such as tilt, through geometric transformation (such as affine, rotation, scaling, etc.), so as to obtain a facial image after aligning facial images at different angles according to the same standard (for example, moving eyebrows, eyes, nose, mouth, facial contours, etc. to the same position).

[0175] (b9) Performing a characterization step: That is, performing facial feature extraction on the facial image after face alignment in step (b8), thereby converting (e.g., encoding) the pixel values ​​of the facial image into features that can represent facial characteristics, thereby obtaining a series of feature vectors of a fixed length to be recognized. Here, in the facial feature extraction process, known architectures such as VGGNet, ResNet, or Transformer can be used.

[0176] (b10) Performing face matching to search for the person: comparing the feature vector to be identified with the facial feature vectors corresponding to the biological identity collected and stored in advance in the face recognition system to search for the person in the face database. Here, when performing the feature vector comparison, the difference between the feature vector to be identified and the pre-stored feature vector can be calculated. If the difference is lower than a preset threshold, the feature vector is considered to be matched; if the difference is higher than the preset threshold, the feature vector is considered to be mismatched. The lower the preset threshold is set, the lower the possibility of misidentification. Instead of using the above-mentioned difference, the judgment can be made by using similarity.

[0177] (b11) Step of drawing a conclusion: Based on the above face matching, if the corresponding personal identity information is found in the face database, a conclusion is drawn, for example, “this is that person”.

[0178] It should be understood that the order of the conversion process, cropping process, and alignment process in the above steps (b6)-(b8) is not particularly limited and can be changed or deleted based on the actual recognition method used. For example, if the image conversion process has been completed before the classification step, this process is no longer required in the recognition stage. For example, the facial key point positioning in the above step (b6) does not have to be completed in step (b6), but can also be completed in step (b3) or step (b5). For example, the image enhancement step is not required.

[0179] In addition, there can usually be a liveness detection step between the above steps (b1) to (b4). For example, the user performs a single action or a combination of actions such as blinking, opening the mouth, shaking the head, and nodding according to the prompt tone. In this way, technologies such as facial key point positioning and face tracking are used to verify whether the user performing the action is a real living person.

[0180] Compared with conventional face recognition using an RGB camera, the face recognition system using this technology can provide more reliable face recognition results.

[0181] [4. Biometric Authentication System of the Fourth Embodiment]

[0182] FIG14 is a block diagram showing a schematic configuration of a biometrics authentication system according to a fourth embodiment of the present disclosure.

[0183] Unlike the previously described biometric authentication system 100C in the third embodiment of the present disclosure, which uses only a SWIR image based on the SWIR spectrum, the biometric authentication system 100D in the fourth embodiment of the present disclosure is further provided with an RGB camera 80 and a synthesis unit 90. The RGB camera 80 generates an RGB image by imaging the visible light portion of the incident light. The synthesis unit 90 outputs a synthesized image by using the SWIR image based on the SWIR spectrum and the RGB image based on the visible light spectrum.

[0184] Specifically, in a biometric recognition system 100D shown in FIG14 , a SWIR camera 10 is used in combination with an RGB camera 80. If the SWIR image captured by the SWIR camera 10 is identified as real skin by the recognition unit 20, the SWIR image is input to the synthesis unit 90. The RGB image captured by the RGB camera 10 is also input to the synthesis unit 90. The synthesis unit 90 synthesizes the SWIR and RGB images by performing a channel superposition operation on the features of interest, and outputs the synthesized image to the recognition unit 60 for recognition.

[0185] As a first modified example of the fourth embodiment of the present disclosure, a depth camera (not shown) or an NIR camera (not shown) may be used instead of the RGB camera 80 .

[0186] Although the fourth embodiment of the present disclosure and its first variation have described an example in which the SWIR camera 10 can be used in combination with any one of an RGB camera, a depth camera, and an NIR camera, as a second variation of the fourth embodiment of the present disclosure, the SWIR camera 10 can be used in combination with any two of these. Furthermore, as a third variation of the fourth embodiment of the present disclosure, the SWIR camera 10 can be used in combination with all three of these.

[0187] It should be noted that the image enhancement function described above as a preferred example in the third embodiment can also be applied to the fourth embodiment of the present disclosure and its various variations. In this case, for example, when referring to FIG. 14 , an image enhancement unit (not shown) can be provided between the synthesis unit 90 and the recognition unit 60, or the image enhancement function can be incorporated into the recognition unit 60 or the synthesis unit 90. This disclosure is not particularly limited to this, as long as the image enhancement process can be performed before the recognition step.

[0188] FIG15 illustrates an operational flow chart of a facial recognition system, which serves as an example of a biometric recognition system according to a third variant of the fourth embodiment of the present disclosure. The operational flow chart of FIG15 differs from the operational flow chart of the third embodiment shown in FIG13 in that face detection and cropping steps are added for images captured by other special cameras, such as an RGB camera, a depth camera, and an NIR camera, as well as an image synthesis step. Descriptions of steps similar to those in FIG13 are omitted here. In the face detection and cropping steps of FIG15 , face detection and ROI frame setting are performed on the SWIR image captured by the SWIR camera, the RGB image captured by the RGB camera, the depth image captured by the depth camera, and the NIR image captured by the NIR camera, respectively, and the ROI frame image is cropped (refer to steps (b1) to (b3) in FIG13 ). In the classification step of FIG15 , a classification model (which can be an AI classification model, a threshold classification model, or a traditional classification model) is trained to distinguish, based at least on the brightness of the SWIR image, whether the living being in the SWIR image is real skin or a mask (refer to step (b4) in FIG13 ). If the classification step identifies the face as a mask, an alarm sounds and an image or text message indicating a "fake face" is output. If the classification step identifies the face as a mask, the image synthesis step (Figure 15) synthesizes the ROI images of the SWIR image, RGB image, depth image, and NIR image, and outputs the synthesized image for recognition. The subsequent recognition process can refer to steps (b5) to (b11) described in Figure 13.

[0189] In the image synthesis step shown in Figure 15, it is not necessary to synthesize all of the SWIR image, RGB image, depth image and NIR image. The image can be output based on one of the SWIR image, RGB image, depth image, and NIR image as needed, or the synthesized image can be output based on any combination of two or more of them.

[0190] The biometric identification method disclosed herein can be implemented with reference to the operational flow chart illustrated in FIG13 or FIG15 , and for example, can include at least some of the steps in the operational flow chart. The biometric identification method includes at least the following identification step: receiving a SWIR image, identifying whether the area corresponding to the SWIR image is real skin or fake skin based at least on the brightness of the SWIR image, and outputting the identification result. The biometric identification method can also include the following SWIR image acquisition step: acquiring the SWIR image for each area of ​​at least one part of the organism based on the SWIR spectrum. The biometric identification method can also optionally include the operational steps or features previously described in the description of the biometric identification system.

[0191] [5. Electronic devices]

[0192] The disclosed technology for distinguishing whether a biological part in a SWIR image is real skin or fake skin based at least on the brightness of the SWIR image can be applied not only to biometric systems and methods but also to electronic devices. For example, the functions of the distinguishing unit, identifying unit, and synthesizing unit described above in the biometric system can all be implemented in code.

[0193] Examples of electronic devices include smartphones with camera functions, facial recognition payment devices, access control systems, and ATM machines. For example, the electronic device includes a memory storing computer-executable code and a processor configured to execute the code to perform the following operations: acquiring or receiving a SWIR image, the SWIR image being acquired based on the SWIR spectrum for each of at least one part of a living being; and determining, based at least on the brightness of the SWIR image, whether the part corresponding to the SWIR image is real skin or fake skin, and outputting the determination result.

[0194] The electronic device may include a SWIR camera for acquiring the SWIR image, or may be connected to a SWIR camera to receive the SWIR image acquired by the SWIR camera. The electronic device may also include or be connected to other special cameras such as an RGB camera, a depth camera, and a NIR camera.

[0195] The electronic device may be configured to further include at least one of a display, a speaker, a microphone, a communication module, a sensor, a touch panel, and a controller, and these components as well as the aforementioned memory, processor, etc. may be connected to each other via a bus.

[0196] When the processor executes the codes, the controller can function as an application processing unit and an operating system processing unit.

[0197] The display can display an operating screen to execute processing performed by the application processing unit and the operating system processing unit, and the touch panel can receive touch operations performed by the user on the operating screen displayed on the display. In addition, the display can also display images, data, and various processing results processed by the electronic device.

[0198] The speaker and microphone can output and collect voice.

[0199] The communication module can perform network communication via a communication network (such as the Internet, a public telephone line network, etc.), a long-distance communication network (such as the so-called 4G lines and 5G lines for wireless mobile bodies, WAN (wide area network), or LAN (local area network), etc.), or a short-range wireless communication (such as Bluetooth (TM), or NFC (near field communication), etc.).

[0200] [6. Application Examples]

[0201] The technology disclosed herein relates to the principles of distinguishing real skin from counterfeit materials used to create fake skin. Therefore, in addition to applications such as facial recognition, this technology can also be applied to technologies for distinguishing fake fingerprints, palm prints, foot prints, or irises. Furthermore, this technology can be applied to any other field requiring identification based on SWIR images captured by a SWIR camera.

[0202] The biometric recognition system disclosed herein uses a SWIR camera, and thus has the following advantages in application examples compared to recognition systems using RGB cameras:

[0203] Enables enhanced material inspection: SWIR cameras can detect chemicals (such as moisture and chemical components) that are not visible in the RGB spectrum. Different substances absorb SWIR light differently, and when captured with a SWIR camera, their images appear different colors. Therefore, SWIR cameras can distinguish between powders of the same color (such as salt, sugar, MSG), liquids (such as water and oil), and plastics, for example, in impurity sorting applications.

[0204] Improved low-light visibility: SWIR cameras perform better in low-light conditions than RGB cameras, thus providing clear images in all lighting conditions.

[0205] Ability to image through obstructions: SWIR cameras can image through obstructions such as fog, smoke, haze, dust, plastic bags, or glass by using the SWIR spectrum, which is difficult for RGB cameras to achieve.

[0206] Ability to inspect food defects based on moisture content: Moisture in food (e.g., soggy bread, bruised fruit, crops with pests) absorbs SWIR light, causing the image to darken. Different moisture contents cause different degrees of darkening. Therefore, by using a SWIR camera to measure the moisture content in food, it is possible to detect whether the food is defective or to classify agricultural products.

[0207] Non-invasive imaging: SWIR cameras can capture images through the surface of the object without intruding or damaging it. This makes them ideal for industrial applications such as detecting defects within wafers and alignment key positions. They are also well-suited for medical and biological applications, such as cosmetic material inspection, as well as non-invasive diagnostics and imaging in dermatology and vascular research.

[0208] It should be noted that the effects described herein are merely illustrative and not restrictive, and the present technology may additionally provide effects other than those described herein.

[0209] It should be understood by those skilled in the art that the embodiments of the present technology are not limited to those described above, and various changes and combinations are possible without departing from the spirit and scope of the present technology.

[0210] It should be noted that the present technology can also adopt the following technical solutions.

[0211] (1) A biometric recognition system comprising a SWIR camera or connected to a SWIR camera and comprising a resolution unit, wherein:

[0212] The SWIR camera collects SWIR images for each of the at least one part of the biological body based on the SWIR spectrum, and

[0213] The distinguishing unit receives the SWIR image captured by the SWIR camera, distinguishes whether the portion corresponding to the SWIR image is real skin or fake skin based on at least the brightness of the SWIR image, and outputs a distinction result.

[0214] (2) The biometric authentication system according to (1) above, wherein:

[0215] The distinguishing unit distinguishes whether the portion corresponding to the SWIR image is real skin or fake skin by means of a classification model.

[0216] (3) The biometric authentication system according to (2) above, wherein:

[0217] The classification model is an AI classification model deployed in the recognition unit after training, which can distinguish whether the part corresponding to the SWIR image is real skin or fake skin based on the brightness of the SWIR image and the features extracted from the SWIR image.

[0218] (4) The biometric authentication system according to (2) above, wherein:

[0219] The classification model is a threshold classification model deployed in the discrimination unit after training, which distinguishes whether the part corresponding to the SWIR image is real skin or fake skin by comparing the brightness of the SWIR image with a preset threshold.

[0220] (5) The biometric authentication system according to (4) above, wherein:

[0221] When the brightness of the SWIR image is equal to or higher than the threshold, the discrimination unit outputs a discrimination result indicating that the portion is fake skin, and

[0222] When the brightness of the SWIR image is lower than the threshold, the discrimination unit outputs a discrimination result indicating that the portion is real skin.

[0223] (6) The biometric recognition system according to any one of (1) to (5) above, further comprising:

[0224] The prompting unit is configured to provide a prompt by generating a specific sound, picture and / or text when the identification result indicates that the part is fake skin.

[0225] (7) The biometric recognition system according to any one of (1) to (6) above, further comprising:

[0226] The recognition unit outputs the SWIR image of the part to the recognition unit when the recognition result indicates that the part is real skin.

[0227] The recognition unit recognizes the identity of the living being in the SWIR image based on image data including at least the SWIR image.

[0228] (8) The biometric authentication system according to (7) above, wherein:

[0229] The recognition unit recognizes the identity of the living being in the SWIR image by matching image data including at least the SWIR image with a pre-stored image data template corresponding to the identity of the living being.

[0230] (9) The biometric recognition system according to (7) or (8) above, further comprising:

[0231] An image enhancement unit performs image enhancement processing on the SWIR image before the recognition unit recognizes the identity of the living being in the SWIR image.

[0232] (10) The biometric authentication system according to any one of (1) to (9), wherein:

[0233] The SWIR camera acquires the SWIR image by using a light source array for generating uniform light as a light source and / or by using an automatic exposure on mode, and / or

[0234] The classification unit performs classification based on brightness and depth information of the SWIR image.

[0235] (11) The biometric recognition system according to any one of (1) to (10) above, further comprising:

[0236] The light source is a halogen lamp.

[0237] (12) The biometric authentication system according to (11) above, wherein:

[0238] A plurality of the halogen lamps are arranged to form a light source array for generating uniform light.

[0239] (13) The biometric recognition system according to any one of (1) to (12) above, further comprising:

[0240] The SWIR bandpass filter is disposed inside the SWIR camera, or is detachably disposed outside the SWIR camera and located between a lens of the SWIR camera and the at least one part of the living being.

[0241] (14) The biometric authentication system according to (13) above, wherein:

[0242] The spectral range of the SWIR bandpass filter is 1400nm to 1600nm.

[0243] (15) The biometric authentication system according to (13) or (14), wherein:

[0244] The spectral range of the SWIR bandpass filter is 1400nm to 1550nm.

[0245] (16) The biometric authentication system according to any one of (13) to (15) above, wherein:

[0246] The spectral range of the SWIR bandpass filter is 1400nm to 1500nm.

[0247] (17) The biometric authentication system according to any one of (13) to (16) above, wherein:

[0248] The spectral range of the SWIR bandpass filter is approximately 1450 nm.

[0249] (18) The biometric authentication system according to any one of (1) to (17) above, wherein:

[0250] The at least one part is at least one of a face, a fingerprint, a foot print, a palm print, and an iris, and

[0251] The fake skin is at least one of a corresponding mask, a fingerprint film, a foot print film, a palm print film, and an iris film.

[0252] (19) The biometric authentication system according to (18) above, wherein:

[0253] When the at least one part is a face, the distinguishing unit performs the following operations:

[0254] performing face detection on the SWIR image from the SWIR camera;

[0255] setting a bounding box containing a face in the SWIR image based on the face detection;

[0256] cropping the SWIR image to retain the image within the bounding box; and

[0257] The image within the bounding box is distinguished.

[0258] (20) The biometric recognition system according to (19) above, further comprising:

[0259] The recognition unit outputs the image within the boundary box to the recognition unit when the recognition result indicates that the part is real skin.

[0260] The recognition unit recognizes the identity of the creature in the image within the bounding box by sequentially performing image enhancement processing, face alignment processing, face representation processing, and face matching processing on the image within the bounding box.

[0261] (21) The biometric authentication system according to any one of (1) to (20) above, wherein:

[0262] The biometric recognition system also includes or is further connected to:

[0263] At least one of an RGB camera, a depth camera, and a NIR camera, which respectively captures an RGB image, a depth image, and a NIR image for each part of the organism based on a corresponding one of a visible light spectrum, distance information, and an NIR spectrum, and

[0264] The biometric recognition system further comprises:

[0265] A synthesis unit, when the discrimination result indicates that the part is real skin, synthesizes the SWIR image with at least one of the RGB image, the depth image, and the NIR image, and outputs the synthesized image for recognition.

[0266] (22) Electronic devices, including:

[0267] a memory storing code executable by a computer; and

[0268] A processor configured to execute the code to perform the following operations:

[0269] receiving SWIR images, wherein the SWIR images are acquired for each of at least one part of the living being based on the SWIR spectrum; and

[0270] Whether the portion corresponding to the SWIR image is real skin or fake skin is determined based at least on the brightness of the SWIR image, and a determination result is output.

[0271] (23) The electronic device according to (22), wherein:

[0272] The wavelength range of the SWIR spectrum is 1400nm to 1600nm.

[0273] (24) The electronic device according to (22) or (23), wherein:

[0274] The SWIR image is acquired by a SWIR camera, and

[0275] The electronic device is connected to the SWIR camera.

[0276] (25) Biometric identification methods, including:

[0277] SWIR image acquisition step: that is, acquiring a SWIR image for each of the at least one part of the biological body based on the SWIR spectrum; and

[0278] The distinguishing step includes receiving the acquired SWIR image, distinguishing whether the portion corresponding to the SWIR image is real skin or fake skin based at least on the brightness of the SWIR image, and outputting the distinguishing result.

[0279] (26) The biometric identification method according to (25) above, further comprising:

[0280] A prompting step is performed using a specific sound, picture and / or text when the identification result indicates that the part is fake skin.

[0281] (27) The biometric identification method according to (25) or (26) above, further comprising:

[0282] Identification step: That is, when the discrimination result indicates that the part is real skin, the identity of the living being in the SWIR image is identified based on image data including at least the SWIR image in the identification step.

[0283] (28) The biometric identification method according to any one of (25) to (27) above, wherein:

[0284] In the SWIR image acquisition step, a light source array for generating uniform light is used as a light source and / or an automatic exposure mode is used, and / or

[0285] In the distinguishing step, the distinguishing is performed based on the brightness and depth information of the SWIR image.

[0286] (29) The biometric identification method according to (27) or (28) above, further comprising:

[0287] Image enhancement step: that is, before performing the recognition in the recognition step, performing image enhancement processing on the SWIR image.

[0288] (30) The biometric identification method according to any one of (25) to (29) above, wherein:

[0289] The wavelength range of the SWIR spectrum is 1400nm to 1600nm.

[0290] (31) The biometric identification method according to any one of (25) to (30) above, further comprising:

[0291] A step of respectively acquiring an RGB image, a depth image, and an NIR image for each part of the organism based on at least one of a visible light spectrum, distance information, and an NIR spectrum; and

[0292] Synthesis step: that is, when the discrimination result indicates that the part is real skin, the SWIR image is synthesized with at least one of the RGB image, the depth image and the NIR image in the synthesis step, and the synthesized image is output for identification.

[0293] [Description of Reference Numerals] 1: Subject 2: Light source 100A, 100B, 100C, 100D: Biometric recognition system 10: SWIR camera 101: Optical element 102: Imaging element (sensor unit) 103: Signal processing unit 105: Display unit 106: Storage unit 20: Resolution unit 201: ROI frame setting unit (face detection unit) 202: ROI frame cropping unit 203: Classification unit 40: Biometric database storage unit 50a: Prompt unit 50b: Monitor 60: Recognition unit 601: Preprocessing unit 602: Alignment unit 603: Representation unit 604: Matching unit 80: RGB camera 90: Synthesis unit

Claims

1. A biometric recognition system comprising a SWIR camera or connected to a SWIR camera and comprising a resolution unit, wherein: The SWIR camera collects SWIR images for each of the at least one part of the biological body based on the SWIR spectrum, and The distinguishing unit receives the SWIR image captured by the SWIR camera, distinguishes whether the portion corresponding to the SWIR image is real skin or fake skin based on at least the brightness of the SWIR image, and outputs a distinction result.

2. The biometric recognition system according to claim 1, wherein: The distinguishing unit distinguishes whether the portion corresponding to the SWIR image is real skin or fake skin by means of a classification model.

3. The biometric identification system according to claim 2, wherein: The classification model is an AI classification model deployed in the recognition unit after training, which can distinguish whether the part corresponding to the SWIR image is real skin or fake skin based on the brightness of the SWIR image and the features extracted from the SWIR image.

4. The biometric recognition system according to claim 2, wherein: The classification model is a threshold classification model deployed in the discrimination unit after training, which distinguishes whether the part corresponding to the SWIR image is real skin or fake skin by comparing the brightness of the SWIR image with a preset threshold.

5. The biometric recognition system according to claim 4, wherein: When the brightness of the SWIR image is equal to or higher than the threshold, the discrimination unit outputs a discrimination result indicating that the portion is fake skin, and When the brightness of the SWIR image is lower than the threshold, the discrimination unit outputs a discrimination result indicating that the portion is real skin.

6. The biometric recognition system according to claim 1, further comprising: The prompting unit is configured to provide a prompt by generating a specific sound, picture and / or text when the identification result indicates that the part is fake skin.

7. The biometric recognition system according to claim 1, further comprising: The recognition unit outputs the SWIR image of the part to the recognition unit when the recognition result indicates that the part is real skin. The recognition unit recognizes the identity of the living being in the SWIR image based on image data including at least the SWIR image.

8. The biometric identification system according to claim 7, wherein: The recognition unit recognizes the identity of the living being in the SWIR image by matching image data including at least the SWIR image with a pre-stored image data template corresponding to the identity of the living being.

9. The biometric recognition system according to claim 7, further comprising: An image enhancement unit performs image enhancement processing on the SWIR image before the recognition unit recognizes the identity of the living being in the SWIR image.

10. The biometric identification system according to claim 1, wherein: The SWIR camera acquires the SWIR image by using a light source array for generating uniform light as a light source and / or by using an automatic exposure on mode, and / or The classification unit performs classification based on brightness and depth information of the SWIR image.

11. The biometric recognition system according to claim 1 , further comprising: The light source is a halogen lamp.

12. The biometric recognition system according to claim 11, wherein: A plurality of the halogen lamps are arranged to form a light source array for generating uniform light.

13. The biometric recognition system according to claim 1, further comprising: The SWIR bandpass filter is disposed inside the SWIR camera, or is detachably disposed outside the SWIR camera and located between a lens of the SWIR camera and the at least one part of the living being.

14. The biometric identification system according to claim 13, wherein: The spectral range of the SWIR bandpass filter is 1400nm to 1600nm.

15. The biometric identification system according to claim 14, wherein: The spectral range of the SWIR bandpass filter is 1400nm to 1550nm.

16. The biometric identification system according to claim 15, wherein: The spectral range of the SWIR bandpass filter is 1400nm to 1500nm.

17. The biometric identification system according to claim 16, wherein: The spectral range of the SWIR bandpass filter is approximately 1450 nm.

18. The biometric identification system according to claim 1, wherein: The at least one part is at least one of a face, a fingerprint, a foot print, a palm print, and an iris, and The fake skin is at least one of a corresponding mask, a fingerprint film, a foot print film, a palm print film, and an iris film.

19. The biometric recognition system according to claim 18, wherein: When the at least one part is a face, the distinguishing unit performs the following operations: performing face detection on the SWIR image from the SWIR camera; setting a bounding box containing a face in the SWIR image based on the face detection; cropping the SWIR image to retain the image within the bounding box; and The image within the bounding box is distinguished.

20. The biometric recognition system according to claim 19, further comprising: The recognition unit outputs the image within the boundary box to the recognition unit when the recognition result indicates that the part is real skin. The recognition unit recognizes the identity of the creature in the image within the bounding box by sequentially performing image enhancement processing, face alignment processing, face representation processing, and face matching processing on the image within the bounding box.

21. The biometric recognition system according to any one of claims 1 to 20, wherein: The biometric recognition system also includes or is further connected to: At least one of an RGB camera, a depth camera, and a NIR camera, which respectively captures an RGB image, a depth image, and a NIR image for each part of the organism based on a corresponding one of a visible light spectrum, distance information, and an NIR spectrum, and The biometric recognition system further comprises: A synthesis unit, when the discrimination result indicates that the part is real skin, synthesizes the SWIR image with at least one of the RGB image, the depth image, and the NIR image, and outputs the synthesized image for recognition.

22. Electronic devices, including: a memory storing code executable by a computer; and A processor configured to execute the code to perform the following operations: receiving SWIR images, wherein the SWIR images are acquired for each of at least one part of the living being based on the SWIR spectrum; and Whether the portion corresponding to the SWIR image is real skin or fake skin is determined based at least on the brightness of the SWIR image, and a determination result is output.

23. The electronic device according to claim 22, wherein: The wavelength range of the SWIR spectrum is 1400nm to 1600nm.

24. The electronic device according to claim 22 or 23, wherein: The SWIR image is acquired by a SWIR camera, and The electronic device includes the SWIR camera or is connected to the SWIR camera.

25. Biometric identification methods, including: SWIR image acquisition step: that is, acquiring a SWIR image for each of the at least one part of the biological body based on the SWIR spectrum; and The distinguishing step includes receiving the acquired SWIR image, distinguishing whether the portion corresponding to the SWIR image is real skin or fake skin based at least on the brightness of the SWIR image, and outputting the distinguishing result.

26. The biometric identification method according to claim 25, further comprising: A prompting step is performed using a specific sound, picture and / or text when the identification result indicates that the part is fake skin.

27. The biometric identification method according to claim 25, further comprising: Identification step: That is, when the discrimination result indicates that the part is real skin, the identity of the living being in the SWIR image is identified based on image data including at least the SWIR image in the identification step.

28. The biometric identification method according to claim 25, wherein: In the SWIR image acquisition step, a light source array for generating uniform light is used as a light source and / or an automatic exposure mode is used, and / or In the distinguishing step, the distinguishing is performed based on the brightness and depth information of the SWIR image.

29. The biometric identification method according to claim 27, further comprising: Image enhancement step: that is, before performing the recognition in the recognition step, performing image enhancement processing on the SWIR image.

30. The biometric identification method according to claim 25, wherein: The wavelength range of the SWIR spectrum is 1400nm to 1600nm.

31. The biometric identification method according to any one of claims 25 to 30, further comprising: A step of respectively acquiring an RGB image, a depth image, and an NIR image for each part of the organism based on at least one of a visible light spectrum, distance information, and an NIR spectrum; and Synthesis step: that is, when the discrimination result indicates that the part is real skin, the SWIR image is synthesized with at least one of the RGB image, the depth image and the NIR image in the synthesis step, and the synthesized image is output for identification.

Images