CN108280418A - The deception recognition methods of face image and device - Google Patents

Abstract

The present disclosure relates to a method and device for deception recognition of facial images, the method comprising: acquiring a facial image to be recognized; using a trained first neural network to extract local features of the face in the facial image, and obtaining The local feature value of the facial image; utilize the trained second neural network to extract the depth feature of the face in the facial image to obtain the depth feature value of the facial image; combine the local feature value and the The depth feature value is fused to obtain the fusion value of the facial image; the fusion value is compared with a threshold, and the deception recognition result of the facial image is judged according to the comparison result. The disclosure has high recognition accuracy and good robustness, and can cope with various image spoofing attacks.

Description

Facial image deception recognition method and device

technical field

The present disclosure relates to the technical field of image recognition, and in particular to a method and device for fraud recognition of facial images.

Background technique

Face recognition technology is a biometric recognition technology based on human facial features. With the continuous development of information technology and the wide application of computer technology, many excellent face recognition algorithms have been produced, such as Fisher face recognition technology. method, local feature analysis method, subspace method, etc., especially after the eigenface method was proposed, face recognition has made further significant development. With the in-depth research on face recognition methods, the current face recognition algorithm has reached a higher level, face recognition has also become the mainstream biometric technology, and has been widely used in practice, such as network account Login, banking system login, access control, face payment, etc. Biometrics technology is closely combined with high-tech means such as computer and optics, acoustics, biosensors and biostatistics, and uses the inherent physiological characteristics of the human body (such as fingerprints, faces, irises, etc.) and behavioral characteristics (such as handwriting, voice, etc.) gait, etc.) for personal identification. Biometrics utilizes physical characteristics such as fingerprints, face and irises, or behavioral characteristics such as typing rhythm and gait to uniquely identify or authenticate individuals. As biometric systems are widely used in real-world applications including mobile phone authentication and access control, biometric spoofing or presentation attacks (PA) are becoming an even greater threat, in which forged biometric samples are presented to biometric Identify the system and attempt to be authenticated. Since the face is the most easily acquired biometric pattern, there are many different types of face recognition for the face including printing attack, replay attack, 3D mask and many more. Therefore, traditional face recognition systems are very vulnerable to such demonstration attacks.

Contents of the invention

In order to overcome the problems existing in related technologies, the present disclosure provides a face image deception recognition method and device, which are used to solve the problem of low face recognition accuracy caused by face image spoofing in traditional methods.

According to an aspect of an embodiment of the present disclosure, there is provided a deception recognition method of a facial image, the method comprising:

Obtain the face image to be recognized;

Using the trained first neural network to extract the local features of the face in the facial image, to obtain the local feature value of the facial image;

Utilize the trained second neural network to extract the depth feature of the face in the face image to obtain the depth feature value of the face image;

Fusing the local feature value and the depth feature value to obtain a fusion value of the face image;

Comparing the fusion value with a threshold, and judging the deception recognition result of the facial image according to the comparison result.

In a possible implementation manner, obtaining the facial image to be recognized includes:

Acquiring a first image and a second image of the face to be recognized, the imaging methods of the first image and the second image are different;

Obtain the facial image to be recognized according to the first image and the second image.

In a possible implementation manner, using the trained first neural network to extract the local features of the face in the face image to obtain the local feature values of the face image, including:

randomly determining a subregion of a face in said face image;

Using the trained first neural network to extract local features in the sub-region to obtain the sub-region features of the face image;

A local feature value of the face image is obtained according to the sub-region features.

In a possible implementation manner, the depth feature of the face in the face image is extracted by using the trained second neural network, and the depth feature value of the face image is obtained, including:

determining a depth calculation area of a face in the face image;

Using the trained second neural network to extract the depth feature of the depth calculation area to obtain the depth feature value of the face image.

In a possible implementation, the fusion value is compared with a threshold, and the deception recognition result of the facial image is judged according to the comparison result, including:

The fusion value is compared with a threshold, and the face image is judged to be a live image or a spoofed image according to the comparison result.

According to another aspect of an embodiment of the present disclosure, there is provided a deception recognition device for facial images, including:

Facial image acquisition module, used to acquire facial images to be identified;

The local feature acquisition module is used to extract the local features of the face in the facial image by using the trained first neural network to obtain the local feature value of the facial image;

The depth feature acquisition module is used to extract the depth feature of the face in the facial image by using the trained second neural network to obtain the depth feature value of the facial image;

A fusion module, configured to fuse the local feature value and the depth feature value to obtain a fusion value of the face image;

The recognition result acquisition module is used to compare the fusion value with a threshold, and judge the deception recognition result of the facial image according to the comparison result.

In a possible implementation manner, the facial image acquisition module includes:

The first image acquisition submodule is used to acquire a first image and a second image of the face to be recognized, and the imaging methods of the first image and the second image are different;

The second image acquisition sub-module is configured to obtain the facial image to be recognized according to the first image and the second image.

In a possible implementation manner, the local feature acquisition module includes:

The sub-region determination submodule is used to randomly determine the sub-region of the face in the facial image;

The sub-area feature acquisition submodule is used to extract local features in the sub-area using the trained first neural network to obtain the sub-area features of the face image;

The local feature acquisition sub-module is used to obtain the local feature value of the facial image according to the sub-region feature.

In a possible implementation, the depth feature acquisition module includes:

A depth calculation area submodule, configured to determine the depth calculation area of the face in the face image;

The depth feature acquisition sub-module is used to extract the depth feature of the depth calculation area by using the trained second neural network, and obtain the depth feature value of the facial image.

In a possible implementation, the recognition result acquisition module includes:

The recognition result acquisition sub-module is used to compare the fusion value with a threshold, and judge that the facial image is a live image or a spoofed image according to the comparison result.

According to another aspect of an embodiment of the present disclosure, there is provided a deception recognition device for facial images, including:

processor;

memory for storing processor-executable instructions;

Wherein, the processor is configured to: execute the method described in any one of the embodiments of the present disclosure.

According to another aspect of the embodiments of the present disclosure, there is provided a non-volatile computer-readable storage medium on which computer program instructions are stored, wherein when the computer program instructions are executed by a processor, the processor can Execute the method described in any one of the embodiments of the present disclosure.

The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: by extracting the local features and depth features of the facial image to be recognized, the fusion feature value of the facial image to be recognized is obtained, and after comparing the fusion feature value with the threshold value, the obtained The deception recognition result of the face image to be recognized. The disclosure has high recognition accuracy and good robustness, and can cope with various image spoofing attacks.

Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings.

Description of drawings

The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the specification, serve to explain the principles of the disclosure.

Fig. 1 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 2 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 3 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 4 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 5 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 6 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 7 is a block diagram of a device for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 8 is a block diagram of a device for identifying fraudulent facial images according to an exemplary embodiment;

Fig. 9 is a block diagram of a device for identifying fraudulent facial images according to an exemplary embodiment.

Fig. 10 is a block diagram of a device for identifying fraudulent facial images according to an exemplary embodiment.

Detailed ways

Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numbers in the figures indicate functionally identical or similar elements. While various aspects of the embodiments are shown in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as superior or better than other embodiments.

In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific implementation manners. It will be understood by those skilled in the art that the present disclosure may be practiced without some of the specific details. In some instances, methods, means, components and circuits that are well known to those skilled in the art have not been described in detail so as to obscure the gist of the present disclosure.

Existing facial image recognition methods include:

1. Texture-based recognition method: The texture features of the face are highly discriminative, so extracting the texture features of the face image can often get a good classification and recognition effect. Image texture feature extraction methods can generally be classified into four categories: statistical methods, model methods, structural methods, and signal processing methods.

2. Based on the way of collecting facial movements: such as the movement of eyeballs and lips, by extracting the facial movements of people in the image to identify whether it is a real face.

3. Correlation method based on image quality and reflectivity: By extracting and comparing the illumination and noise information on the image to determine whether it is a real face.

However, there is no great correlation between the pixel density of the extracted texture features and different attack methods, so it is very difficult to extract stable texture features. The method of collecting facial motion has certain advantages for recognizing static images, but for Video or image replay attacks are ineffective. The method based on image quality and reflectivity has high requirements on the image and is sensitive to noise, which is not conducive to stable recognition.

Fig. 1 is a flow chart of a deception recognition method for facial images according to an exemplary embodiment. As shown in Fig. 1, the deception recognition method for facial images may include:

Step S10, acquiring a face image to be recognized.

Step S20, using the trained first neural network to extract the local features of the face in the face image to obtain the local feature values of the face image.

Step S30, using the trained second neural network to extract the depth features of the face in the face image to obtain the depth feature value of the face image.

In step S40, the local feature value and the depth feature value are fused to obtain a fusion value of the facial image.

Step S50, comparing the fusion value with a threshold value, and judging the deception recognition result of the facial image according to the comparison result.

The facial images to be recognized may include multiple types. For example, a facial image of a living body such as a person captured on the spot, and a face image of a non-living body that may be a spoofing image.

Since the local features of the face are extracted separately, it is not conducive to resisting the attack of the existing spoofing images. The disclosure divides the face in the facial image to be recognized into multiple sub-regions, and extracts the local features of each sub-region of the facial image to be recognized. The disclosure also extracts the overall depth feature of the face image to be recognized by referring to the depth of the face itself in the face image to be recognized. The disclosure fuses the extracted local features and overall features, compares the fusion value with a set threshold, and judges whether the facial image to be recognized is a deceptive image according to the comparison result, thereby improving the recognition accuracy of the deceptive image.

Fig. 2 is a flow chart of a face image deception recognition method shown according to an exemplary embodiment. As shown in Fig. 2, the difference from the above embodiment is that step S10 includes:

Step S11, acquiring a first image and a second image of the face to be recognized, where the imaging methods of the first image and the second image are different.

Step S12, obtaining the face image to be recognized according to the first image and the second image.

In a possible implementation manner, the first image is acquired by means of infrared imaging, and the first image is an infrared light image. The second image is acquired by means of visible light imaging, and the second image is a visible light image. Image fusion is performed on the first image and the second image to obtain a facial image to be recognized. Image fusion is divided into three levels from low to high: data-level fusion, feature-level fusion and decision-level fusion.

Among them, data-level fusion, also known as pixel-level fusion, refers to the process of directly processing the data collected by the sensor to obtain a fused image, which can keep as much original data as possible on site. In feature-level fusion, it can be guaranteed that different images contain information features, such as the representation of infrared light for object heat, and the representation of visible light for object brightness. Decision-level fusion mainly lies in subjective requirements, and some rules can be used, such as Bayesian method, D-S evidence method and voting method.

Face recognition is performed on the fused image using image recognition technology, including face detection using a trained face recognition neural network to obtain a face image.

Fig. 3 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment. As shown in Fig. 3, the difference from the above embodiment is that step S20 includes:

Step S21, randomly determining a face sub-region in the face image.

Step S22, using the trained first neural network to extract local features in the sub-region to obtain the sub-region features of the face image.

Step S23, according to the sub-region feature, obtain the local feature value of the face image.

In a possible implementation, the face subregions randomly determined in the face image to be recognized may include eye subregions, nose subregions, mouth subregions, etc. The sub-regions may also include sub-regions smaller than each part of the face. Use the trained first neural network to extract local features in each sub-region, such as extracting SIFT (Scale-invariant feature transform, scale-invariant feature transform) features, LBP (Local Binary Patterns, local binary pattern features) features, HOG (Histogram of OrientedGradient, histogram of oriented gradient) feature. After data processing is performed on the extracted local features of each sub-region, local feature values of the face image to be recognized are obtained.

Fig. 4 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment. As shown in Fig. 4, the difference from the above embodiment is that step S30 includes:

Step S31, determining the depth calculation area of the face in the face image.

Step S32, using the trained second neural network to extract the depth features of the depth calculation area to obtain the depth feature values of the face image.

In a possible implementation, in the present disclosure, the depth refers to the depth of the face region with reference to a certain position of itself, not the depth of other objects, nor the distance from the face region to other external positions. distance. To calculate the depth of the face image, it is not necessary to calculate all the regions in the face image. One way is to first determine the depth calculation area in the face image, and only calculate the depth of the face in the depth calculation area. In the determined depth calculation area, after using the second neural network to extract the depth feature, the depth feature value of the facial image is obtained according to the extracted depth feature.

Fig. 5 is a flow chart of a method for identifying fraudulent facial images according to an exemplary embodiment. As shown in Fig. 5, the difference from the above embodiment is that step S50 includes:

Step S51 , comparing the fusion value with a threshold, and judging that the facial image is a live image or a spoofed image according to the comparison result.

In a possible implementation manner, the fusion value is obtained after the local feature value and the depth feature value of the face image are fused. Then compare the fusion value with the set threshold. If the fusion value is higher than the threshold, the face image to be recognized is a spoofed image. If the fusion value is lower than the threshold, the face image to be recognized is a live image.

In order to better illustrate the method of the present disclosure, the following example is an exemplary embodiment of the present disclosure. Fig. 6 is a flow chart of a method for identifying fraudulent facial images according to another exemplary embodiment, as shown in Fig. 6 , including:

Step 1, using different imaging methods to acquire images respectively. For example, an infrared light image is captured by an infrared camera, and a visible light image is captured by a visible light camera. In practical applications, binocular cameras can be used to simultaneously acquire infrared light images and visible light images.

Step 2, after fusing the infrared light image and the visible light image, an input image is obtained. Refer to related descriptions in the foregoing embodiments.

Step 3, after detecting the human face in the input image, obtain the facial image for analysis. This includes face recognition in incoming images using image recognition technology. Refer to related descriptions in the foregoing embodiments.

In step 4, the facial images are respectively input into two neural networks, and the upper and lower two processes are used to identify the processing steps in two CNNs (Convolutional Neural Networks, Convolutional Neural Networks) in Fig. 6 . In Figure 6, the CNN pipeline in the upper part processes the local features of the face image. Extract local patch features in the face image, that is, after randomly dividing the sub-regions in the face image, extract the local features of each sub-region respectively. In the CNN process in the lower part, the face image is input into the depth-based CNN to extract the overall depth features of the face image.

As there are more and more ways to perceive the environment and image deception, feature extraction alone for identification can no longer cover all attacks. Therefore, convolutional neural networks are used to learn from massive data, and use massive training data to distinguish between the scene and deception sample. For patch-based CNN, a deep convolutional neural network is trained end-to-end to learn rich appearance features, capable of distinguishing live and off-site face images using randomly extracted patches from face images. For depth-based CNNs, a fully convolutional network (FCN) is trained to estimate the depth of face images, assuming that print or replay image attacks have flat depth maps, while live faces contain normal facial depth. And depth-based CNNs can independently detect facial attacks based on appearance or depth cues.

Step 5. In the CNN process in the upper part, the extracted local special diagnosis of each sub-region is input into the CNN based on local patches for processing. In the CNN process in the lower part, the depth features of the face image are obtained according to the depth features output by the depth-based CNN. The deep features utilize the whole face and describe the face as a 3D object, while the non-living face is described as a flat plane.

Estimating depth from a single RGB image is a fundamental problem in computer vision. For face images, face reconstruction from an image or multiple images can be considered as a means of depth estimation. The present disclosure estimates the depths of live and non-living faces, using the face itself as a depth reference instead of calculating the fixed distance between the face and the camera that captures the face as the depth.

Step 6, in the CNN process in the upper part, estimate the activity score for each local feature, so as to obtain the score of the local feature of the face image. In the CNN process in the lower part, SVM (Support Vector Machine, Support Vector Machine) classification is performed on the overall feature to obtain the score of the deep feature of the face image.

The CNN in the upper part is trained end-to-end and assigns each randomly extracted patch score from face images. The average score of face images is then assigned. The CNN in the lower part estimates the depth map of the face image and provides liveness scores to the face image based on the estimated depth map.

Step 7, after fusing the scores of the local features and the scores of the deep features, compare the fusion score with the set threshold.

Step 8, according to the comparison result, it is judged that the facial image is a fake image or a live image. The fusion output of local feature score and deep feature score is called spoof score. If the spoof score is higher than a predefined threshold, the face image or video clip is classified as a non-live image.

This embodiment proposes a new method of using two convolutional neural networks for face verification and anti-spoofing: one neural network is used to extract local features, one neural network is used to extract deep features, and the depth map obtained by extracting independent global features to verify whether the input image is an entity. This embodiment is based on the construction of a two-way neural network to separately identify local features and overall depth features, and the technology of fusion and comparison of local features and depth features to distinguish scores can realize the recognition of living faces faster and more conveniently, and detect whether it is true before identity verification. It is a living body and can effectively detect spoofing attacks such as photos, videos, and 3D masks.

This embodiment collects images simultaneously based on an infrared camera and a visible light camera, and uses a two-way neural network for local and depth discrimination. The technology is novel, the recognition accuracy is high, and the robustness is good, which can cope with various image spoofing attacks.

Fig. 7 is a block diagram of a deception recognition device for facial images according to an exemplary embodiment. As shown in Fig. 7, the deception recognition device for facial images includes:

Facial image acquisition module 61, used to acquire facial images to be identified;

The local feature acquisition module 62 is used to extract the local feature of the face in the facial image by using the trained first neural network to obtain the local feature value of the facial image;

Depth feature acquisition module 63, for utilizing the trained second neural network to extract the depth feature of the face in the facial image, to obtain the depth feature value of the facial image;

A fusion module 64, configured to fuse the local feature value and the depth feature value to obtain a fusion value of the face image;

The recognition result acquisition module 65 is configured to compare the fusion value with a threshold, and judge the deception recognition result of the facial image according to the comparison result.

Fig. 8 is a block diagram of a face image deception recognition device according to an exemplary embodiment, as shown in Fig. 8 ,

In a possible implementation manner, the facial image acquisition module 61 includes:

The first image acquisition sub-module 611 is configured to acquire a first image and a second image of the face to be recognized, and the imaging methods of the first image and the second image are different;

The second image acquisition sub-module 612 is configured to obtain the face image to be recognized according to the first image and the second image.

In a possible implementation, the local feature acquisition module 62 includes:

The sub-region determining sub-module 621 is used to randomly determine the sub-region of the face in the facial image;

The sub-area feature acquisition submodule 622 is used to extract local features in the sub-area by using the trained first neural network to obtain the sub-area features of the face image;

The local feature acquisition sub-module 623 is used to obtain the local feature value of the face image according to the sub-region feature.

In a possible implementation, the depth feature acquisition module 63 includes:

The depth calculation area submodule 631 is used to determine the depth calculation area of the face in the facial image;

The depth feature acquisition sub-module 632 is configured to use the trained second neural network to extract the depth features of the depth calculation area to obtain the depth feature value of the facial image.

In a possible implementation, the recognition result acquisition module 65 includes:

The recognition result acquisition sub-module 651 is configured to compare the fusion value with a threshold, and judge that the facial image is a live image or a spoofed image according to the comparison result.

Fig. 9 is a block diagram of an apparatus 800 for deception recognition of facial images according to an exemplary embodiment. For example, the apparatus 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, and the like.

Referring to FIG. 9, the device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and communication component 816 .

The processing component 802 generally controls the overall operations of the device 800, such as those associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above method. Additionally, processing component 802 may include one or more modules that facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802 .

The memory 804 is configured to store various types of data to support operations at the device 800 . Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, images, video, and the like. The memory 804 can be implemented by any type of volatile or non-volatile storage device or their combination, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic or Optical Disk.

The power supply component 806 provides power to the various components of the device 800 . Power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for device 800 .

The multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may not only sense a boundary of a touch or swipe action, but also detect duration and pressure associated with the touch or swipe action. In some embodiments, the multimedia component 808 includes a front camera and/or a rear camera. When the device 800 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a microphone (MIC) configured to receive external audio signals when the device 800 is in operation modes, such as call mode, recording mode and voice recognition mode. Received audio signals may be further stored in memory 804 or sent via communication component 816 . In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and a peripheral interface module, which may be a keyboard, a click wheel, a button, and the like. These buttons may include, but are not limited to: a home button, volume buttons, start button, and lock button.

Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of device 800 . For example, the sensor component 814 can detect the open/closed state of the device 800, the relative positioning of components, such as the display and keypad of the device 800, and the sensor component 814 can also detect a change in the position of the device 800 or a component of the device 800 , the presence or absence of user contact with the device 800 , the device 800 orientation or acceleration/deceleration and the temperature change of the device 800 . Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. Sensor assembly 814 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor or a temperature sensor.

The communication component 816 is configured to facilitate wired or wireless communication between the apparatus 800 and other devices. The device 800 can access wireless networks based on communication standards, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology and other technologies.

In an exemplary embodiment, apparatus 800 may be programmed by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A gate array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation for performing the methods described above.

In an exemplary embodiment, there is also provided a non-volatile computer-readable storage medium, such as the memory 804 including computer program instructions, which can be executed by the processor 820 of the device 800 to implement the above method.

Fig. 10 is a block diagram of an apparatus 1900 for deception recognition of facial images according to an exemplary embodiment. For example, apparatus 1900 may be provided as a server. Referring to FIG. 10 , apparatus 1900 includes processing component 1922 , which further includes one or more processors, and a memory resource represented by memory 1932 for storing instructions executable by processing component 1922 , such as application programs. The application programs stored in memory 1932 may include one or more modules each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

Device 1900 may also include a power component 1926 configured to perform power management of device 1900 , a wired or wireless network interface 1950 configured to connect device 1900 to a network, and an input-output (I/O) interface 1958 . The apparatus 1900 can operate based on an operating system stored in the memory 1932, such as Windows Server™, MacOS X™, Unix™, Linux™, FreeBSD™ or the like.

In an exemplary embodiment, there is also provided a non-transitory computer-readable storage medium, such as the memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the apparatus 1900 to implement the above-mentioned method.

The present disclosure can be a system, method and/or computer program product. A computer program product may include a computer readable storage medium having computer readable program instructions thereon for causing a processor to implement various aspects of the present disclosure.

A computer readable storage medium may be a tangible device that can retain and store instructions for use by an instruction execution device. A computer readable storage medium may be, for example, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory), static random access memory (SRAM), compact disc read only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanically encoded device, such as a printer with instructions stored thereon A hole card or a raised structure in a groove, and any suitable combination of the above. As used herein, computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., pulses of light through fiber optic cables), or transmitted electrical signals.

Computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or downloaded to an external computer or external storage device over a network, such as the Internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers. A network adapter card or a network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .

Computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or Source or object code written in any combination, including object-oriented programming languages—such as Smalltalk, C++, etc., and conventional procedural programming languages—such as the “C” language or similar programming languages. Computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as via the Internet using an Internet service provider). connect). In some embodiments, an electronic circuit, such as a programmable logic circuit, field programmable gate array (FPGA), or programmable logic array (PLA), can be customized by utilizing state information of computer-readable program instructions, which can Various aspects of the present disclosure are implemented by executing computer readable program instructions.

Aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It should be understood that each block of the flowcharts and/or block diagrams, and combinations of blocks in the flowcharts and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine such that when executed by the processor of the computer or other programmable data processing apparatus , producing an apparatus for realizing the functions/actions specified in one or more blocks in the flowchart and/or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause computers, programmable data processing devices and/or other devices to work in a specific way, so that the computer-readable medium storing instructions includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks in flowcharts and/or block diagrams.

It is also possible to load computer-readable program instructions into a computer, other programmable data processing device, or other equipment, so that a series of operational steps are performed on the computer, other programmable data processing device, or other equipment to produce a computer-implemented process , so that instructions executed on computers, other programmable data processing devices, or other devices implement the functions/actions specified in one or more blocks in the flowcharts and/or block diagrams.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, a portion of a program segment, or an instruction that includes one or more Executable instructions. In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It should also be noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by a dedicated hardware-based system that performs the specified function or action , or may be implemented by a combination of dedicated hardware and computer instructions.

Having described various embodiments of the present disclosure above, the foregoing description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and alterations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles of the various embodiments, practical applications or technical improvements over technologies in the market, or to enable other persons of ordinary skill in the art to understand the various embodiments disclosed herein.

Claims (12)

1. A method for deceiving recognition of facial images, characterized in that the method comprises: Obtain the face image to be recognized; Using the trained first neural network to extract the local features of the face in the facial image, to obtain the local feature value of the facial image; Utilize the trained second neural network to extract the depth feature of the face in the face image to obtain the depth feature value of the face image; Fusing the local feature value and the depth feature value to obtain a fusion value of the facial image; Comparing the fusion value with a threshold, and judging the deception recognition result of the facial image according to the comparison result. 2. The method according to claim 1, wherein obtaining the facial image to be identified comprises: Acquiring a first image and a second image of the face to be recognized, the imaging methods of the first image and the second image are different; Obtain the facial image to be recognized according to the first image and the second image. 3. The method according to claim 1, wherein, utilizing the trained first neural network to extract the local feature of the face in the facial image, obtain the local feature value of the facial image, comprising: randomly determining a subregion of a face in said face image; Using the trained first neural network to extract local features in the sub-region to obtain the sub-region features of the face image; A local feature value of the face image is obtained according to the sub-region features. 4. The method according to claim 1, wherein, utilizing the trained second neural network to extract the depth feature of the face in the facial image, obtain the depth feature value of the facial image, comprising: determining a depth calculation area of a face in the face image; Using the trained second neural network to extract the depth feature of the depth calculation area to obtain the depth feature value of the face image. 5. The method according to claim 1, characterized in that, comparing the fusion value with a threshold, and judging the deception recognition result of the face image according to the comparison result, comprising: The fusion value is compared with a threshold, and the face image is judged to be a live image or a spoofed image according to the comparison result. 6. A deception recognition device for facial images, characterized in that it comprises: Facial image acquisition module, used to acquire facial images to be identified; The local feature acquisition module is used to extract the local features of the face in the facial image by using the trained first neural network to obtain the local feature value of the facial image; The depth feature acquisition module is used to extract the depth feature of the face in the facial image by using the trained second neural network to obtain the depth feature value of the facial image; A fusion module, configured to fuse the local feature value and the depth feature value to obtain a fusion value of the face image; The recognition result acquisition module is used to compare the fusion value with a threshold, and judge the deception recognition result of the facial image according to the comparison result. 7. The device according to claim 6, wherein the facial image acquisition module includes: The first image acquisition submodule is used to acquire a first image and a second image of the face to be recognized, and the imaging methods of the first image and the second image are different; The second image acquisition sub-module is configured to obtain the facial image to be recognized according to the first image and the second image. 8. The device according to claim 6, wherein the local feature acquisition module comprises: The sub-region determination submodule is used to randomly determine the sub-region of the face in the facial image; The sub-area feature acquisition submodule is used to extract local features in the sub-area using the trained first neural network to obtain the sub-area features of the face image; The local feature acquisition sub-module is used to obtain the local feature value of the facial image according to the sub-region feature. 9. The device according to claim 6, wherein the depth feature acquisition module comprises: A depth calculation area submodule, configured to determine the depth calculation area of the face in the face image; The depth feature acquisition sub-module is used to extract the depth feature of the depth calculation area by using the trained second neural network, and obtain the depth feature value of the facial image. 10. The device according to claim 6, wherein the recognition result acquisition module includes: The recognition result acquisition sub-module is used to compare the fusion value with a threshold, and judge that the facial image is a live image or a spoofed image according to the comparison result. 11. A deception recognition device for facial images, characterized in that it comprises: processor; memory for storing processor-executable instructions; Wherein, the processor is configured to: execute the method described in any one of claims 1-5. 12. A non-volatile computer-readable storage medium, on which computer program instructions are stored, wherein when the computer program instructions are executed by a processor, the processor can perform any one of the methods described.