Method, apparatus, and computer readable storage medium for liveness detection

Abstract

The application discloses a method, device and computer readable storage medium for living body detection. The method comprises the following steps: acquiring a thermal imaging image of a target object, and dividing the thermal imaging image into multiple regions; determining the region temperature of each region in the thermal imaging image; performing feature extraction based on the region temperatures of the multiple regions to obtain a temperature distribution feature of the target object; and performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body. The method provided by the application can reduce the algorithm calculation amount in the living body detection process and improve the speed of living body detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a living body detection method and device and computer readable storage medium. BACKGROUND

[0002] The development of digital technology brings human society into the era of artificial intelligence, and various biometric technologies are widely used in people's daily life. With the development of biometric technologies based on human faces, they are widely used in user identity authentication systems such as gate access, mobile phone unlocking, and attendance systems. However, these face recognition authentication systems have huge security problems. For example, when the user's face information is stolen, it is put on the mobile phone screen, printed into a photo, or made into a mask to disguise the user's face for identification and authentication, and attacks the identity authentication system. If the face authentication system does not accurately detect the disguised face information, it will cause immeasurable loss to the user, so it is necessary to detect the living body of the face to be authenticated.

[0003] Face living body detection technology is a technology that analyzes face image information to determine whether the face obtained by the camera is a disguised attack face or a real living body face. However, the current face living body detection technology has the problems of slow detection speed and low detection efficiency. SUMMARY

[0004] The present application provides a living body detection method, device and computer readable storage medium, which can improve the speed of living body detection.

[0005] The first aspect of the embodiment of the present application provides a living body detection method, which comprises: acquiring a thermal image of a target object, and dividing the thermal image into a plurality of regions; determining the region temperature of each region in the thermal image; performing feature extraction based on the region temperature of each region to obtain the temperature distribution feature of the target object; and performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body.

[0006] The second aspect of the embodiment of the present application provides a living body detection device, which comprises: an acquisition module for acquiring a thermal image of a target object and dividing the thermal image into a plurality of regions; a determination module connected to the acquisition module for determining the region temperature of each region in the thermal image; an extraction module connected to the determination module for performing feature extraction based on the region temperature of each region to obtain the temperature distribution feature of the target object; and a detection module connected to the extraction module for performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body.

[0007] The third aspect of the embodiment of the present application provides a living body detection device, the living body detection device comprises a processor, a memory and a communication circuit, the processor is coupled with the memory and the communication circuit respectively, the memory stores program data, and the processor realizes the steps in the above method by executing the program data in the memory.

[0008] The fourth aspect of the embodiment of the present application provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program can be executed by a processor to realize the steps in the above method.

[0009] Beneficial effects are that the present application divides the thermal imaging image into multiple regions, extracts the temperature distribution characteristics of the target object based on the region temperature of the multiple regions, and reduces the subsequent calculation amount by dividing the thermal imaging image into regions and realizing the down-sampling processing of the image data in the processing process, thereby improving the speed of living body detection. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0011] Figure 1 is a flowchart of an embodiment of the method for living body detection of the present application;

[0012] Figure 2 is Figure 1 a flowchart of step S120 in

[0013] Figure 3 is Figure 1 a flowchart of step S140 in

[0014] Figure 4 is Figure 1 a framework diagram of the method for living body detection in an application scenario;

[0015] Figure 5 is a structural diagram of an embodiment of the living body detection device of the present application;

[0016] Figure 6 is a structural diagram of another embodiment of the living body detection device of the present application;

[0017] Figure 7 is a structural diagram of an embodiment of the computer readable storage medium of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of the present application.

[0019] Referring to Figure 1 , Figure 1 is a flowchart of an embodiment of a method for living body detection, and the method comprises the following steps.

[0020] S110: Obtain a thermal imaging image of a target object, and divide the thermal imaging image into multiple regions.

[0021] The thermal imaging image comprises the target object, and the purpose of living body detection is to determine whether the target object in the thermal imaging image is a living body, that is, the target object is an object to be subjected to living body authentication. The thermal imaging image of the target object is obtained by a thermal imaging infrared camera photographing the target object. After the thermal imaging infrared camera photographs the target object, the image output by the thermal imaging infrared camera can be directly used as the thermal imaging image of the target object, or a detection frame that frames the target object can be determined in the thermal imaging image, and then the image in the detection frame is used as the thermal imaging image of the target object. The process of determining the detection frame that frames the target object in the thermal imaging image can be referred to below.

[0022] In the embodiment, in order to reduce the amount of calculation and improve the speed of living body detection, the thermal imaging image is divided into multiple regions to perform down-sampling processing on the image data of the thermal imaging image.

[0023] The sizes of the multiple regions can be the same or different, but in the embodiment, the sizes of the multiple regions are set to be the same, that is, the step S110 equally divides the thermal imaging image into multiple regions with the same size. The length and width of each region can be the same or different, which is not limited herein.

[0024] In an application scenario, the thermal imaging image is divided into i rows and i columns of regions, for example, the thermal imaging image is divided into 9 regions (that is, 3 rows and 3 columns of regions), 16 regions (that is, 4 rows and 4 columns of regions), or 25 regions (that is, 5 rows and 5 columns of regions).

[0025] S120: Determine the region temperature of each region in the thermal imaging image respectively.

[0026] The thermal imaging image carries temperature information, because the region temperature of each region can be determined after the regions are divided.

[0027] Referring to Figure 2 In this embodiment, the step S120 of determining the region temperature of each region comprises:

[0028] S121: respectively determining the average temperature value of the pixel points in each region.

[0029] Specifically, in the thermal imaging image, each pixel point has a temperature value, so after the regions are divided, for each region, the average temperature value T of the pixel points is calculated using the following formula:

[0030] Wherein, N is the number of pixel points in the region, t i represents the temperature value of the i-th pixel point in the region.

[0031] S122: respectively determining the average temperature value of each region as the region temperature of each region.

[0032] Specifically, after step S121, each region has a corresponding average temperature value, so the average temperature value of each region is respectively determined as the region temperature of each region.

[0033] That is, in this embodiment, the average temperature value of the pixel points in the region is determined as the region temperature of the region, but the present application is not limited thereto, and in other embodiments, the maximum temperature value, the minimum temperature value or the median temperature value of the pixel points in the region can also be determined as the region temperature. In summary, the present application does not limit the specific process of determining the region temperature of the region.

[0034] S130: performing feature extraction based on the region temperatures of the multiple regions to obtain the temperature distribution feature of the target object.

[0035] Specifically, after the above steps, each region has a region temperature, and then feature extraction is performed based on the region temperatures of all regions to obtain the temperature distribution feature of the target object.

[0036] In this embodiment, step S130 specifically comprises: performing convolution operation on the thermal imaging image based on the region temperatures of the multiple regions to obtain the temperature distribution feature of the target object.

[0037] Wherein, the thermal imaging image can be input into a convolutional neural network, and the convolutional neural network performs convolution operation based on the region temperatures of all regions, so as to extract the high-dimensional temperature feature of the thermal imaging image, i.e. the temperature distribution feature of the target object.

[0038] S140: performing live body detection based on the temperature distribution feature of the target object to determine whether the target object is a live body.

[0039] In an application scenario, when the temperature distribution feature of the target object matches the temperature distribution feature of a living body, it is determined that the target object is a living body, otherwise it is determined that the target object is a non-living body.

[0040] In another application scenario, as shown in FIG. 14, step S140 specifically includes: Figure 3

[0041] S141: respectively performing living body detection feature extraction on the at least one target image to obtain the living body detection feature of the target object, wherein the at least one target image includes at least one of a color image, a near-infrared image and a thermal imaging image of the target object, and the living body detection feature of the target object is different from the temperature distribution feature of the target object.

[0042] Specifically, the living body detection feature of the target object is different from the temperature distribution feature of the target object, which means that the living body detection feature of the target object is not obtained by temperature feature extraction on the target object. For example, when the living body detection feature of the target object is obtained by performing living body detection feature extraction on the near-infrared image, since the near-infrared image cannot display the temperature information of the target object, the extracted living body detection feature naturally cannot contain temperature information, and the obtained living body detection feature is naturally different from the temperature distribution feature.

[0043] Alternatively, the living body detection feature of the target object is obtained by temperature feature extraction on the target object, but the extraction process of the living body detection feature of the target object is different from the process of obtaining the temperature distribution feature. For example, when performing living body detection feature extraction on the thermal imaging image, the process of performing living body detection feature extraction on the thermal imaging image is set to be different from the process of obtaining the temperature distribution feature, so as to ensure that the living body detection feature is different from the temperature distribution feature.

[0044] The living body detection feature of the target object can represent the confidence that the target object is a living body. In related technologies, living body detection is directly performed based on the living body detection feature of the target object, so as to determine whether the target object is a living body. However, in this embodiment, in order to improve the accuracy of living body detection, living body detection is not directly performed based on the living body detection feature of the target object, and the living body detection feature of the target object needs to be fused with the obtained temperature distribution feature in the future. For details, please refer to the following.

[0045] The color image of the target object is obtained by shooting the target object by a white light camera, and the near-infrared image of the target object is obtained by shooting the target object by a near-infrared camera.

[0046] In an application scenario, the step of obtaining the color image, the near-infrared image and the thermal imaging image of the target object includes:​

[0047] (a) obtaining a first image, a second image and a third image by respectively shooting a target object by a white light camera, a near-infrared camera and a thermal imaging infrared camera.

[0048] (b) performing target recognition on the first image to obtain a first position of the target object in the first image, and extracting a color image of the target object from the first image according to the first position.

[0049] Specifically, the first image is obtained by shooting with the white light camera, and the first image has high definition, so the accuracy of target recognition is high.

[0050] The first position of the target object in the first image can be the position of the detection box of the target object in the first image, for example, the first position includes the coordinates of the top-left vertex and the bottom-right vertex of the detection box of the target object in the first image.

[0051] The size of the finally obtained color image of the target object is smaller than the size of the first image, and the proportion of the target object in the final color image is greater than the proportion of the target object in the first image.

[0052] (c) according to the first position, respectively marking the near-infrared image and the thermal imaging image of the target object in the second image and the third image.

[0053] Specifically, according to the first position of the target object in the first image and the relative installation position between the white light camera and the near-infrared camera, the second position of the target object in the second image can be determined, and then according to the second position, the near-infrared image of the target object can be extracted from the second image. It can be understood that at this time, the proportion of the target object in the finally obtained near-infrared image is greater than the proportion of the target object in the second image. Wherein, the position of the target object in the second image can be the position of the detection box of the target object in the second image.

[0054] Similarly, according to the first position of the target object in the first image and the relative installation position between the white light camera and the thermal imaging infrared camera, the third position of the target object in the third image can be determined, and then according to the third position, the thermal imaging image of the target object can be extracted from the third image. It can be understood that at this time, the proportion of the target object in the finally obtained thermal imaging image is greater than the proportion of the target object in the third image. Wherein, the position of the target object in the third image can be the position of the detection box of the target object in the third image.

[0055] In the above scheme, on the one hand, considering that the accuracy of target recognition on the first image captured by the white light camera is high, therefore, based on the position of the target object in the first image, the accuracy of extracting the near-infrared image and the thermal image of the target object from the second image and the third image is high, on the other hand, directly according to the first position, the near-infrared image and the thermal image of the target object are extracted from the second image and the third image respectively, without the need to separately identify the target in the second image and the third image, which can reduce the calculation amount of the algorithm, and further improve the speed of the entire detection method.

[0056] It should be noted that in other embodiments, the second image can also be directly identified to determine the position of the target object, and then the near-infrared image of the target object is extracted from the second image based on the position of the target object. Similarly, the third image can also be directly identified to determine the position of the target object, and then the near-infrared image of the target object is extracted from the third image based on the position of the target object.

[0057] Alternatively, in other embodiments, only the second image or the third image can be identified, for example, only the second image is identified to determine the second position of the target object in the second image, and then the near-infrared image of the target object is extracted from the second image according to the second position, and the color image and the thermal image of the target object are marked in the first image and the third image based on the second position.

[0058] Alternatively, in other embodiments, the first image can also be directly used as the color image of the target object, or the second image can be directly used as the near-infrared image of the target object, or the third image can be directly used as the thermal image of the target object.

[0059] In summary, the specific process of determining the color image, the near-infrared image and the thermal image of the target object is not limited in the present application.

[0060] Continuing to refer to Figure 3 In step S141, one of the color image, the near-infrared image and the thermal image can be subjected to living body detection feature extraction to directly obtain the living body detection feature of the target object, for example, the near-infrared image is subjected to living body detection feature extraction to obtain the living body detection feature of the target object.

[0061] Alternatively, in step S141, multiple (for example, two or three) of the color image, the near-infrared image and the thermal image can be subjected to living body detection feature extraction respectively, and then the living body detection feature of the target object is obtained. For example, the color image and the near-infrared image are subjected to living body detection feature extraction respectively to obtain the living body detection feature of the target object.

[0062] In an application scenario, after performing live body detection feature extraction on multiple ones of the color image, the near-infrared image, and the thermal imaging image, live body detection features of each image are obtained, and then the live body detection features corresponding to all the images are fused to obtain live body detection features of the target object. Any fusion technology in the prior art can be used to fuse the live body detection features corresponding to all the images, and the specific process of the fusion is not described in detail here.

[0063] That is, step S141 includes: performing live body detection feature extraction on at least one target image respectively to obtain live body detection features corresponding to each target image; and fusing the live body detection features corresponding to all the target images to obtain live body detection features of the target object. It can be understood that when live body detection feature extraction is performed on only one target image, the live body detection features of the target image are directly taken as the live body detection features of the target object, and when live body detection feature extraction is performed on multiple target images, live body detection features of each target image are obtained, and then the live body detection features corresponding to the multiple target images are fused to obtain live body detection features of the target object.

[0064] The process of performing live body detection feature extraction on the target image includes: performing feature extraction on the target image by using a pre-trained live body detection network to obtain live body detection features of the target image.

[0065] Specifically, when the live body detection network is used to perform feature extraction on the target image, different types of images use different live body detection networks. For example, the live body detection network used for performing live body detection feature extraction on the color image, the live body detection network used for performing live body detection feature extraction on the near-infrared image, and the live body detection network used for performing live body detection feature extraction on the thermal imaging image are three different live body detection networks, which are defined as a first live body detection network, a second live body detection network, and a third live body detection network, respectively. In the training stage, when the first live body detection network is trained, the sample image used is an image captured by a white light camera, when the second live body detection network is trained, the sample image used is an image captured by a near-infrared camera, and when the third live body detection network is trained, the sample image used is an image captured by a thermal imaging camera.

[0066] S142: Fusing the temperature distribution features and the live body detection features to obtain fused features.

[0067] In an application scenario, the temperature distribution feature and the living body detection feature are spliced to obtain a fusion feature. For example, the size of the temperature distribution feature is 1*10*9*9 (1 represents a dimension, and 10 is a channel number), and the size of the living body detection feature is 1*12*9*9 (1 represents a dimension, and 12 is a channel number). After splicing, the size of the fusion feature is 1*(10+12)*9*9 (1 represents a dimension).

[0068] The temperature distribution feature and the living body detection feature can be fused by using any fusion method in the prior art to obtain a fusion feature. The specific process of the fusion is not described here.

[0069] S143: Living body detection is performed based on the fusion feature to determine whether the target object is a living body.

[0070] Specifically, since the fusion feature contains both the temperature distribution feature of the target object and the living body detection feature of the target object, compared with living body detection based only on the temperature distribution feature of the target object or living body detection based only on the living body detection feature of the target object, the embodiment can improve the accuracy of living body detection.

[0071] In an application scenario, in order to further improve the accuracy and speed of detection, a pre-trained neural network is used for living body detection. Specifically, the fusion feature is input into the neural network to determine whether the target object is a living body.

[0072] In the related art, the temperature distribution feature of the thermal imaging image is generally extracted directly based on the temperature values of all pixel points in the thermal imaging image. In the embodiment, the target object is divided into multiple regions, and the temperature distribution feature of the target object is extracted based on the region temperature of the multiple regions. In the processing, the image data is down-sampled by dividing the thermal imaging image into regions, so that the subsequent calculation amount can be reduced, and the speed of living body detection can be improved.

[0073] In order to better understand the scheme of the present application, the scheme of the present application will be described in detail below in combination with the application scenario of Figure 4 .

[0074] First, a first image, a second image, and a third image obtained by capturing the target object by using a white light camera, a near-infrared camera, and a thermal imaging infrared camera are acquired, respectively. Then, target recognition is performed on the first image to obtain a first position of the target object in the first image. Then, according to the first position, a near-infrared image and a thermal imaging image of the target object are marked in the second image and the third image, respectively.

[0075] Then, the near-infrared image (in the second image) and the thermal imaging image (in the third image) are input into a neural network to obtain a first feature and a second feature, respectively. Figure 4In the diagram, the near-infrared image of the target object (denoted by label 101) is input into a pre-trained liveness detection network to obtain the liveness detection features of the target object (in...). Figure 4 In the text, the liveness detection features of the target object are denoted by label 201.

[0076] And thermal imaging images ( Figure 4 In the process, the thermal imaging image (represented by label 102) is divided into multiple regions, and the average temperature value of the pixels in each region is determined. Finally, the average temperature value of the pixels in each region is determined as the region temperature of each region.

[0077] Based on the regional temperatures of multiple areas, a convolution operation is performed on the thermal imaging image to obtain the temperature distribution characteristics of the target object (in... Figure 4 In the text, the temperature distribution characteristics of the target object are represented by the symbol 202.

[0078] Next, the liveness detection features and temperature distribution features of the target object are fused to obtain the fused features (in Figure 4 In the text, the fusion feature is indicated by the symbol 20.

[0079] Finally, liveness detection is performed based on the fused features to determine whether the target object is alive. For example, the fused features are incorporated into the fully connected layer of a pre-trained neural network for calculation to determine whether the target object is alive.

[0080] See Figure 5 , Figure 5 This is a schematic diagram of one embodiment of the liveness detection device of this application. The liveness detection device 200 includes a processor 210, a memory 220, and a communication circuit 230. The processor 210 is coupled to the memory 220 and the communication circuit 230. The memory 220 stores program data. The processor 210 executes the program data in the memory 220 to implement the steps in any of the above embodiments. Detailed steps can be found in the above embodiments and will not be repeated here.

[0081] The liveness detection device 200 can be any device with image processing capabilities, such as a computer or mobile phone, and there are no restrictions on its use.

[0082] See Figure 6 , Figure 6 This is a schematic diagram of one embodiment of the liveness detection device of this application. The liveness detection device 300 includes an acquisition module 310, a determination module 320, an extraction module 330, and a detection module 340.

[0083] The acquisition module 310 is used to acquire thermal imaging images of the target object and divide the thermal imaging images into multiple regions.

[0084] The determining module 320 is connected with the acquiring module 310, and is configured to determine the region temperature of each region in the thermal imaging image.

[0085] The extracting module 330 is connected with the determining module 320, and is configured to perform feature extraction based on the region temperatures of the multiple regions to obtain a temperature distribution feature of the target object.

[0086] The detecting module 340 is connected with the extracting module 330, and is configured to perform living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body.

[0087] The living body detection apparatus 300 performs the steps in the method for living body detection in any one of the above embodiments when working, and the detailed steps can be referred to the above related content, which will not be repeated here.

[0088] The living body detection apparatus 300 can be any device with image processing capability such as a computer or a mobile phone, which is not limited here.

[0089] Referring to Figure 7 , Figure 7 is a structural schematic diagram of an embodiment of the computer readable storage medium of the present application. The computer readable storage medium 400 stores a computer program 410, and the computer program 410 can be executed by a processor to implement the steps in any one of the above methods.

[0090] The computer readable storage medium 400 can be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or any device that can store the computer program 410, or can be a server that stores the computer program 410, which can send the stored computer program 410 to other devices for running, or can run the stored computer program 410 itself.

[0091] The above is only an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method of detecting living entities, characterized in that, The method comprises: acquiring a thermal imaging image of a target object and dividing the thermal imaging image into a plurality of regions; the plurality of regions have the same size; determining a region temperature of each of the regions in the thermal imaging image respectively; the step of determining the region temperature of each of the regions in the thermal imaging image respectively comprises: determining one of an average temperature value, a maximum temperature value, a minimum temperature value and a median temperature value of each of the regions respectively; and determining the one of the average temperature value, the maximum temperature value, the minimum temperature value and the median temperature value of each of the regions as the region temperature of each of the regions respectively; performing feature extraction based on the region temperatures of the plurality of regions to obtain a temperature distribution feature of the target object; the step of performing feature extraction based on the region temperatures of the plurality of regions to obtain the temperature distribution feature of the target object comprises: performing a convolution operation on the thermal imaging image based on the region temperatures of the plurality of regions to obtain the temperature distribution feature of the target object; performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body; the step of performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body comprises: performing living body detection feature extraction on at least one target image respectively to obtain a living body detection feature corresponding to each of the target images respectively; performing fusion processing on the living body detection features corresponding to all the target images to obtain a living body detection feature of the target object, wherein the at least one target image comprises at least one of a color image, a near-infrared image and the thermal imaging image of the target object; performing fusion processing on the temperature distribution feature and the living body detection feature to obtain a fusion feature; and performing living body detection based on the fusion feature to determine whether the target object is a living body.

2. The method of claim 1, wherein, the step of performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body comprises: performing living body detection feature extraction on at least one target image respectively to obtain a living body detection feature of the target object, wherein the at least one target image comprises at least one of a color image, a near-infrared image and the thermal imaging image of the target object, and the living body detection feature of the target object is different from the temperature distribution feature of the target object; performing fusion processing on the temperature distribution feature and the living body detection feature to obtain a fusion feature; performing living body detection based on the fusion feature to determine whether the target object is a living body.

3. The method of claim 2, wherein, when the at least one target image comprises the near-infrared image of the target object, the method further comprises, before the step of acquiring the thermal imaging image of the target object: acquiring a first image, a second image and a third image obtained by respectively photographing the target object by a white light camera, a near-infrared camera and a thermal imaging infrared camera; performing target recognition on the first image to obtain a first position of the target object in the first image; According to the first position, the near-infrared image and the thermal imaging image of the target object are calibrated in the second image and the third image, respectively.

4. A living body detecting apparatus characterized by comprising: The living body detection device comprises: An acquisition module is configured to acquire a thermal imaging image of a target object and divide the thermal imaging image into a plurality of regions, wherein the plurality of regions have the same size; A determination module is connected to the acquisition module and configured to determine a region temperature of each region in the thermal imaging image, respectively; The step of determining the region temperature of each region in the thermal imaging image, respectively, comprises: determining one of an average temperature value, a maximum temperature value, a minimum temperature value, and a median temperature value of each pixel point in each region, respectively; and determining the one of the average temperature value, the maximum temperature value, the minimum temperature value, and the median temperature value of each region as the region temperature of each region, respectively. An extraction module is connected to the determination module and configured to perform feature extraction based on the region temperatures of the plurality of regions to obtain a temperature distribution feature of the target object; The step of performing feature extraction based on the region temperatures of the plurality of regions to obtain the temperature distribution feature of the target object comprises: performing a convolution operation on the thermal imaging image based on the region temperatures of the plurality of regions to obtain the temperature distribution feature of the target object. A detection module is connected to the extraction module and configured to perform living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body. The step of performing living body detection based on the temperature distribution feature of the target object to determine whether the target object is a living body comprises: performing living body detection feature extraction on at least one target image to obtain a living body detection feature corresponding to each target image, respectively; performing fusion processing on the living body detection features corresponding to all target images to obtain a living body detection feature of the target object, wherein the at least one target image comprises at least one of a color image, a near-infrared image, and a thermal imaging image of the target object; performing fusion processing on the temperature distribution feature and the living body detection feature to obtain a fusion feature; and performing living body detection based on the fusion feature to determine whether the target object is a living body.

5. A living body detecting apparatus characterized by comprising: The living body detection device comprises a processor, a memory, and a communication circuit, the processor is coupled to the memory and the communication circuit, respectively, the memory stores program data, and the processor executes the program data in the memory to realize the steps in the method of any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that can be executed by a processor to realize the steps in the method of any one of claims 1-3.

Images