WO2022213955A1 - 图像采集设备劫持的检测方法、装置及计算机设备 - Google Patents

图像采集设备劫持的检测方法、装置及计算机设备 Download PDF

Info

Publication number
WO2022213955A1
WO2022213955A1 PCT/CN2022/085189 CN2022085189W WO2022213955A1 WO 2022213955 A1 WO2022213955 A1 WO 2022213955A1 CN 2022085189 W CN2022085189 W CN 2022085189W WO 2022213955 A1 WO2022213955 A1 WO 2022213955A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
acquisition device
image acquisition
distance
hijacking
Prior art date
Application number
PCT/CN2022/085189
Other languages
English (en)
French (fr)
Inventor
刘宇光
何果财
吕军
裴积全
王帅廷
Original Assignee
京东科技控股股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 京东科技控股股份有限公司 filed Critical 京东科技控股股份有限公司
Publication of WO2022213955A1 publication Critical patent/WO2022213955A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition

Definitions

  • the present application relates to the technical field of information processing, and in particular, to a method, device and computer equipment for detecting hijacking of an image acquisition device.
  • the face image sequence recorded in advance can be used to replace the image sequence actually captured by the image capture device, so as to achieve the purpose of face attack. For example, if A wants to attack B's account, A can obtain B's personal image sequence and photos, perform image sequence synthesis and other processing, obtain B's face recognition image sequence, and then modify the data interface of the image acquisition device of its own device. When the sequence is shot and uploaded, B's face recognition image sequence is used to replace the image sequence actually shot by the image acquisition device, and then uploaded to realize the attack on B.
  • the present application aims to solve one of the technical problems in the related art at least to a certain extent.
  • the present application proposes a detection method, device and computer equipment for hijacking of an image acquisition device, so as to realize that different shooting parameters in the image sequence actually shot by the image acquisition device correspond to different images, and different shooting parameters in the hijacking of the image acquisition device correspond to different images There will be anomalies in the image, so that the synthetic image sequence and the real image sequence can be distinguished, the accuracy of the image sequence detection is improved, and the detection accuracy of whether the image acquisition device is hijacked can be improved.
  • An embodiment of the first aspect of the present application provides a method for detecting hijacking of an image acquisition device, including:
  • a hijacking detection result of the image acquisition device is generated.
  • a shooting instruction is sent to the image acquisition device, so that the image acquisition device performs a shooting operation; during the shooting process of the image acquisition device, the shooting parameters of the image acquisition device are dynamically adjusted. , and send the dynamically adjusted shooting parameters to the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters to capture images; acquire the image sequence captured by the image acquisition device;
  • different shooting parameters in the image sequence actually captured by the image acquisition device can correspond to different images, and when different shooting parameters correspond to different images in the hijacking of the image acquisition device, abnormality will occur, so that the synthetic image sequence can be distinguished from the real image sequence.
  • the image sequence obtained by shooting improves the detection efficiency of the image sequence, thereby improving the detection accuracy of whether the image acquisition device is hijacked.
  • the embodiment of the second aspect of the present application provides a detection device for hijacking of an image acquisition device, including:
  • a sending module configured to send a shooting instruction to an image acquisition device, so that the image acquisition device performs a shooting operation
  • a dynamic adjustment module configured to dynamically adjust the shooting parameters of the image acquisition device during the shooting process of the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters for shooting;
  • an acquisition module configured to acquire an image sequence captured by the image acquisition device
  • a generating module is configured to generate a hijacking detection result of the image acquisition device according to the image sequence.
  • a shooting instruction is sent to the image acquisition device, so that the image acquisition device performs a shooting operation; during the shooting process of the image acquisition device, the shooting parameters of the image acquisition device are dynamically adjusted. , and send the dynamically adjusted shooting parameters to the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters to capture images; acquire the image sequence captured by the image acquisition device;
  • different shooting parameters in the image sequence actually captured by the image acquisition device can correspond to different images, and when different shooting parameters correspond to different images in the hijacking of the image acquisition device, abnormality will occur, so that the synthetic image sequence can be distinguished from the real image sequence.
  • the image sequence obtained by shooting improves the detection efficiency of the image sequence, thereby improving the detection accuracy of whether the image acquisition device is hijacked.
  • the embodiment of the third aspect of the present application proposes a computer device, including: a memory, a processor, and a computer program stored in the memory and running on the processor.
  • the processor executes the program, the computer program as described in the present application
  • a method for detecting hijacking of an image acquisition device provided by an embodiment of the first aspect.
  • Embodiments of the fourth aspect of the present application provide a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to cause the computer to perform the detection of hijacking of an image acquisition device provided by the embodiments of the first aspect of the present application method.
  • the embodiment of the fifth aspect of the present application provides a computer program product.
  • an instruction processor in the computer program product is executed, the method for detecting hijacking of an image acquisition device provided by the embodiment of the first aspect of the present application is executed.
  • FIG. 1 is a schematic flowchart of a method for detecting hijacking of an image acquisition device according to Embodiment 1 of the present application;
  • FIG. 2 is a schematic structural diagram of a detection device for hijacking of an image acquisition device provided in Embodiment 2 of the present application;
  • Figure 3 shows a block diagram of an exemplary computer device suitable for use in implementing embodiments of the present application.
  • the server can detect whether the uploaded image sequence is a composite image sequence through a detection algorithm for the composite image sequence, and then determine whether the uploaded image sequence is an image sequence actually shot by an image acquisition device.
  • the detection algorithm of the synthesized image sequence is difficult to be universally used in various image sequence synthesis algorithms, resulting in poor detection accuracy of the synthesized image sequence, and then whether the image acquisition equipment is hijacked. The detection accuracy is poor.
  • the present application mainly aims at the technical problems of poor detection accuracy of synthetic image sequences in the prior art and poor detection accuracy of image acquisition equipment hijacking, and proposes a detection method for image acquisition equipment hijacking.
  • a shooting instruction is sent to the image acquisition device, so that the image acquisition device performs a shooting operation; during the shooting process of the image acquisition device, the shooting parameters of the image acquisition device are dynamically adjusted. , and send the dynamically adjusted shooting parameters to the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters to capture images; acquire the image sequence captured by the image acquisition device;
  • different shooting parameters in the image sequence actually captured by the image acquisition device can correspond to different images, and when different shooting parameters correspond to different images in the hijacking of the image acquisition device, abnormality will occur, so that the synthetic image sequence can be distinguished from the real image sequence.
  • the image sequence obtained by shooting improves the detection efficiency of the image sequence, thereby improving the detection accuracy of whether the image acquisition device is hijacked.
  • FIG. 1 is a schematic flowchart of a method for detecting hijacking of an image acquisition device according to Embodiment 1 of the present application.
  • the embodiment of the present application is illustrated by taking the method for detecting hijacking of an image acquisition device configured in an apparatus for detecting hijacking of an image acquisition device.
  • the detection function of image acquisition device hijacking can be performed.
  • the computer equipment can be a personal computer (Personal Computer, PC for short), a cloud device, a mobile device, etc.
  • the mobile device can be, for example, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, a vehicle-mounted device, etc. with various operating systems, Hardware devices for touch screens and/or display screens.
  • the method for detecting hijacking of an image acquisition device may include the following steps S101 to S104.
  • Step 101 Send a shooting instruction to an image acquisition device, so that the image acquisition device performs a shooting operation.
  • the triggering condition for the detection device hijacked by the image capture device to send the shooting instruction to the image capture device may be, for example, the user of the image capture device is logging in to a bank account or transferring money, and needs to perform face liveness detection; another example may be, Need to detect whether the image acquisition device is hijacked.
  • Step 102 During the shooting process of the image acquisition device, dynamically adjust the shooting parameters of the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters for shooting.
  • the photographing process of the image acquisition device may be, for example, using the original photographing parameters to photograph to obtain the first image; receiving the dynamic adjustment instruction of the photographing parameters, adjusting the original photographing parameters, obtaining the adjusted photographing parameters, and adopting the adjustment
  • a second image is obtained by shooting with the following shooting parameters, and the above process is repeated to obtain an image sequence.
  • the shooting parameters of the image acquisition device may include at least one of the following parameters: exposure time, exposure compensation, exposure gain, focal length, resolution, shutter parameters, aperture parameters, whether to turn on the flash, and the like.
  • the dynamic adjustment strategy of the image capturing device hijacking detection device to the shooting parameters of the image capturing device may be a preset adjustment strategy or a random adjustment, and may be set according to actual needs.
  • Step 103 Acquire an image sequence captured by the image acquisition device.
  • Step 104 Generate a hijacking detection result of the image acquisition device according to the image sequence.
  • the process of performing step 104 by the detection device for hijacking of the image acquisition device may specifically be: for each adjacent image pair in the image sequence, input the adjacent image pair into the hijacking detection model to generate adjacent image pairs The distance between the two images in ; the hijacking detection result is determined according to the distance corresponding to each adjacent image pair.
  • the hijacking detection model includes: a dimensionality reduction processing layer and a distance detection layer connected in sequence; wherein, the dimensionality reduction processing layer is used to perform dimensionality reduction processing on two images in an adjacent image pair, so that image acquisition When the device is not hijacked, the distance between the two images after dimensionality reduction processing is as small as possible. When the image acquisition device is hijacked, the distance between the two images after dimensionality reduction processing is as large as possible.
  • the training implementation of the detection model; the distance detection layer is used to detect the distance between the two images after dimensionality reduction processing.
  • the distance here may refer to the Euclidean distance between two images, and the Euclidean distance may represent the similarity between the two images.
  • the image after dimensionality reduction processing may be specifically represented by a vector, and the dimension of the vector may be, for example, 128 dimensions or the like.
  • the training process of the hijacking detection model can be, for example, obtaining the initial hijacking detection model; obtaining training data, wherein the training data includes: the first sample image under the first shooting parameters, and the image acquisition device is not hijacked The second sample image under the second shooting parameter when the image acquisition device is hijacked and the third sample image under the second shooting parameter when the image acquisition device is hijacked; A distance, and the second distance between the first sample image and the third sample image are output, and an objective function is constructed by the difference between the first distance and the second distance, and the coefficient of the initial hijacking detection model is determined according to the objective function value. Make adjustments to achieve training.
  • the objective function can be specifically shown in the following formula (1).
  • W and b represent the coefficients of the hijacking detection model, represents the first sample image, represents the second sample image, represents the third sample image, f represents the dimensionality reduction algorithm, represents the first distance, Represents the second distance, a represents the minimum target interval between the first distance and the second distance; the "+" sign indicates that when the value in square brackets is greater than zero, the objective function value is the value in square brackets; the value in square brackets When less than or equal to zero, the objective function value is zero.
  • the above objective function can be used to make the difference between the second distance and the first distance infinitely close to the value of a in the output of the hijacking detection model.
  • the distance between the two images should be as small as possible.
  • the distance between the two images after dimensionality reduction processing should be as large as possible, so that the distance judgment can be used to determine whether the image sequence is a composite image sequence or a composite image sequence.
  • the process of determining the hijacking detection result according to the distance corresponding to each adjacent image pair by the device for detecting hijacking of the image acquisition device may be, for example, when there are adjacent images whose corresponding distance is greater than or equal to a preset distance threshold When the time is correct, it is determined that the image acquisition device is hijacked; or, when the proportion of adjacent image pairs whose corresponding distances in the image sequence are greater than or equal to the preset distance threshold is greater than or equal to the preset proportion threshold, determine the image acquisition device hijacked.
  • the proportion of adjacent image pairs whose corresponding distance is greater than or equal to the preset distance threshold in the image sequence is greater than or equal to the preset proportion threshold, it is determined that the image acquisition device is hijacked, considering that the image acquisition device captures the image sequence During the process, the shooting parameters are adjusted according to the change of the position of the shooting object and the change of the light at the position, and the proportion of adjacent image pairs whose corresponding distance is greater than or equal to the preset distance threshold in the image sequence is greater than or equal to the preset proportion.
  • the ratio is higher than the threshold, it is determined that the image acquisition device is hijacked, which can further improve the detection accuracy of whether the image acquisition device is hijacked.
  • the method may further include the following steps: when the hijacking detection result is that the image acquisition device is hijacked, face recognition processing is not performed on the image sequence; when the hijacking detection result is that the image acquisition device is hijacked When not hijacked, face recognition processing is performed on the image sequence to perform face liveness detection, etc.
  • a shooting instruction is sent to the image acquisition device, so that the image acquisition device performs a shooting operation; during the shooting process of the image acquisition device, the shooting parameters of the image acquisition device are dynamically adjusted. , and send the dynamically adjusted shooting parameters to the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters to capture images; acquire the image sequence captured by the image acquisition device;
  • different shooting parameters in the image sequence actually captured by the image acquisition device can correspond to different images, and when different shooting parameters correspond to different images in the hijacking of the image acquisition device, abnormality will occur, so that the synthetic image sequence can be distinguished from the real image sequence.
  • the image sequence obtained by shooting improves the detection efficiency of the image sequence, thereby improving the detection accuracy of whether the image acquisition device is hijacked.
  • FIG. 2 is a schematic structural diagram of an apparatus for detecting hijacking of an image acquisition device according to Embodiment 2 of the present application.
  • the apparatus 200 for detecting hijacking of an image acquisition device may include: a sending module 210 , a dynamic adjustment module 220 , an obtaining module 230 and a generating module 240 .
  • the sending module 210 is configured to send a shooting instruction to an image acquisition device, so that the image acquisition device performs a shooting operation;
  • a dynamic adjustment module 220 configured to dynamically adjust the shooting parameters of the image acquisition device during the process of shooting by the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters to shoot;
  • an acquisition module 230 configured to acquire an image sequence captured by the image acquisition device
  • the generating module 240 is configured to generate a hijacking detection result of the image acquisition device according to the image sequence.
  • the generating module 240 is specifically configured to:
  • the hijacking detection result is determined.
  • the hijacking detection model includes: a dimensionality reduction processing layer and a distance detection layer connected in sequence;
  • the dimensionality reduction processing layer is used to perform dimensionality reduction processing on two images in the adjacent image pair;
  • the distance detection layer is used to detect the distance between the two images after dimension reduction processing.
  • the device further includes: a training module
  • the obtaining module 230 is further configured to obtain an initial hijacking detection model
  • the acquisition module 230 is further configured to acquire training data, wherein the training data includes: a first sample image under the first shooting parameter, and a second sample image under the second shooting parameter when the image acquisition device is not hijacked and the third sample image under the second shooting parameter when the image acquisition device is hijacked;
  • the training module is configured to take the training data as input, take the first distance between the first sample image and the second sample image, and the first sample image and the third sample image
  • the second distance between the sample images is output, an objective function is constructed based on the difference between the first distance and the second distance, and the coefficients of the initial hijacking detection model are adjusted according to the objective function value to realize training.
  • the objective function is,
  • W and b represent the coefficients of the hijacking detection model, represents the first sample image, represents the second sample image, represents the third sample image, f represents the dimensionality reduction algorithm, represents the first distance,
  • a represents the minimum target separation between the first distance and the second distance.
  • the generating module 240 is specifically configured to determine that the image acquisition device is hijacked when there are adjacent image pairs whose corresponding distances are greater than or equal to a preset distance threshold; When the proportion of adjacent image pairs whose corresponding distance is greater than or equal to the preset distance threshold in the image sequence is greater than or equal to the preset proportion threshold, it is determined that the image acquisition device is hijacked.
  • the dynamic adjustment module is specifically configured to randomly and dynamically adjust the shooting parameters of the image acquisition device.
  • the apparatus further includes: a processing module, configured to not perform face recognition processing on the image sequence when the hijacking detection result is that the image acquisition device is hijacked; When the hijacking detection result is that the image acquisition device is not hijacked, face recognition processing is performed on the image sequence.
  • a shooting instruction is sent to the image acquisition device, so that the image acquisition device performs a shooting operation; during the shooting process of the image acquisition device, the shooting parameters of the image acquisition device are dynamically adjusted. , and send the dynamically adjusted shooting parameters to the image acquisition device, so that the image acquisition device uses the dynamically adjusted shooting parameters to capture images; acquire the image sequence captured by the image acquisition device;
  • different shooting parameters in the image sequence actually captured by the image acquisition device can correspond to different images, and when different shooting parameters correspond to different images in the hijacking of the image acquisition device, abnormality will occur, so that the synthetic image sequence can be distinguished from the real image sequence.
  • the image sequence obtained by shooting improves the detection efficiency of the image sequence, thereby improving the detection accuracy of whether the image acquisition device is hijacked.
  • the present application can also propose a detection system for hijacking of an image acquisition device, including: a terminal device and a cloud device; wherein, the terminal device is provided with an image acquisition device, and the cloud device can be connected with the terminal device, and the execution is as above
  • a detection service may be set on the cloud device, or a detection service on other devices may be called to generate a hijacking detection result of the image acquisition device according to the image sequence.
  • the present application also proposes a computer device, including: a memory, a processor, and a computer program stored in the memory and running on the processor.
  • a computer program stored in the memory and running on the processor.
  • the present application also proposes a computer program product, when the instruction processor in the computer program product executes, executes the detection method for hijacking of an image acquisition device as proposed in the foregoing embodiments of the present application.
  • Figure 3 shows a block diagram of an exemplary computer device suitable for use in implementing embodiments of the present application.
  • the computer device 12 shown in FIG. 3 is only an example, and should not impose any limitations on the functions and scope of use of the embodiments of the present application.
  • computer device 12 takes the form of a general-purpose computing device.
  • Components of computer device 12 may include, but are not limited to, one or more processors or processing units 16 , system memory 28 , and a bus 18 connecting various system components including system memory 28 and processing unit 16 .
  • Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of a variety of bus structures.
  • these architectures include, but are not limited to, Industry Standard Architecture (hereinafter referred to as: ISA) bus, Micro Channel Architecture (hereinafter referred to as: MAC) bus, enhanced ISA bus, video electronics Standards Association (Video Electronics Standards Association; hereinafter referred to as: VESA) local bus and Peripheral Component Interconnection (Peripheral Component Interconnection; hereinafter referred to as: PCI) bus.
  • ISA Industry Standard Architecture
  • MAC Micro Channel Architecture
  • VESA Video Electronics Standards Association
  • PCI Peripheral Component Interconnection
  • Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including both volatile and nonvolatile media, removable and non-removable media.
  • the memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (Random Access Memory; hereinafter: RAM) 30 and/or cache memory 32 .
  • Computer device 12 may further include other removable/non-removable, volatile/non-volatile computer system storage media.
  • storage system 34 may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 3, commonly referred to as a "hard disk drive").
  • magnetic disk drives for reading and writing to removable non-volatile magnetic disks eg "floppy disks" and removable non-volatile optical disks (eg, compact disk read only memory) may be provided Disc Read Only Memory; hereinafter referred to as: CD-ROM), Digital Video Disc Read Only Memory (hereinafter referred to as: DVD-ROM) or other optical media) read and write optical drives.
  • CD-ROM Disc Read Only Memory
  • DVD-ROM Digital Video Disc Read Only Memory
  • each drive may be connected to bus 18 through one or more data media interfaces.
  • Memory 28 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present application.
  • Program modules 42 generally perform the functions and/or methods of the embodiments described herein.
  • Computer device 12 may also communicate with one or more external devices 14 (eg, keyboard, pointing device, display 24, etc.), may also communicate with one or more devices that enable a user to interact with computer device 12, and/or communicate with Any device (eg, network card, modem, etc.) that enables the computer device 12 to communicate with one or more other computing devices. Such communication may take place through input/output (I/O) interface 22 .
  • the computer device 12 can also communicate with one or more networks (such as a local area network (Local Area Network; hereinafter referred to as: LAN), a wide area network (Wide Area Network; hereinafter referred to as: WAN) and/or a public network, such as the Internet, through the network adapter 20 ) communication.
  • networks such as a local area network (Local Area Network; hereinafter referred to as: LAN), a wide area network (Wide Area Network; hereinafter referred to as: WAN) and/or a public network, such as the Internet, through the network
  • network adapter 20 communicates with other modules of computer device 12 via bus 18 .
  • bus 18 It should be understood that, although not shown, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives and data backup storage systems.
  • the processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28 , for example, implements the methods mentioned in the foregoing embodiments.
  • first and second are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature delimited with “first”, “second” may expressly or implicitly include at least one of that feature.
  • plurality means at least two, such as two, three, etc., unless expressly and specifically defined otherwise.
  • a "computer-readable medium” can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or apparatus.
  • computer readable media include the following: electrical connections with one or more wiring (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), Read Only Memory (ROM), Erasable Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM).
  • the computer readable medium may even be paper or other suitable medium on which the program may be printed, as the paper or other medium may be optically scanned, for example, followed by editing, interpretation, or other suitable medium as necessary process to obtain the program electronically and then store it in computer memory.
  • each functional unit in each embodiment of the present application may be integrated into one processing module, or each unit may exist physically alone, or two or more units may be integrated into one module.
  • the above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.
  • the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, and the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Studio Devices (AREA)

Abstract

本申请提出一种图像采集设备劫持的检测方法、装置及计算机设备，其中方法包括：向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作；在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调整，并将动态调整后的拍摄参数发送给图像采集设备，以使图像采集设备采用动态调整后的拍摄参数拍摄图像；获取图像采集设备拍摄得到的图像序列；从而能够使得图像采集设备真实拍摄得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，最终生成图像采集设备的劫持检测结果。

Description

图像采集设备劫持的检测方法、装置及计算机设备
相关申请的交叉引用
本申请基于申请号为202110369263.X、申请日为2021年04月06日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及信息处理技术领域，尤其涉及一种图像采集设备劫持的检测方法、装置及计算机设备。
背景技术
目前，存在一种人脸攻击方式，通过修改图像采集设备数据接口，可实现采用提前录制好的人脸图像序列，替换图像采集设备真实拍摄到的图像序列，达到人脸攻击的目的。例如，A想攻击B的账户，A可获取B的个人图像序列和照片，进行图像序列合成等处理，获取B的人脸识别图像序列，然后修改自己设备的图像采集设备数据接口，在进行图像序列拍摄并上传时，采用B的人脸识别图像序列替换图像采集设备实际拍摄的图像序列，然后上传，实现对B的攻击。
发明内容
本申请旨在至少在一定程度上解决相关技术中的技术问题之一。
本申请提出一种图像采集设备劫持的检测方法、装置及计算机设备，以实现图像采集设备真实拍摄得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，提高图像序列检测的准确度，提高图像采集设备是否劫持的检测准确度。
本申请第一方面实施例提出了一种图像采集设备劫持的检测方法，包括：
向图像采集设备发送拍摄指令，以使所述图像采集设备进行拍摄操作；
在所述图像采集设备拍摄的过程中，对所述图像采集设备的拍摄参数进行动态调整，以使所述图像采集设备采用动态调整后的拍摄参数拍摄图像；
获取所述图像采集设备拍摄得到的图像序列；
根据所述图像序列，生成所述图像采集设备的劫持检测结果。
本申请实施例的图像采集设备劫持的检测方法，通过向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作；在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调整，并将动态调整后的拍摄参数发送给图像采集设备，以使图像采集设备采用动态调整后的拍摄参数拍摄图像；获取图像采集设备拍摄得到的图像序列；根据图像序列，生成图像采集设备的劫持检测结果，从而能够使得图像采集设备真实拍摄 得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，提高图像序列的检测效率，进而提高图像采集设备是否劫持的检测准确度。
本申请第二方面实施例提出了一种图像采集设备劫持的检测装置，包括：
发送模块，用于向图像采集设备发送拍摄指令，以使所述图像采集设备进行拍摄操作；
动态调整模块，用于在所述图像采集设备拍摄的过程中，对所述图像采集设备的拍摄参数进行动态调整，以使所述图像采集设备采用动态调整后的拍摄参数进行拍摄；
获取模块，用于获取所述图像采集设备拍摄得到的图像序列；
生成模块，用于根据所述图像序列，生成所述图像采集设备的劫持检测结果。
本申请实施例的图像采集设备劫持的检测装置，通过向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作；在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调整，并将动态调整后的拍摄参数发送给图像采集设备，以使图像采集设备采用动态调整后的拍摄参数拍摄图像；获取图像采集设备拍摄得到的图像序列；根据图像序列，生成图像采集设备的劫持检测结果，从而能够使得图像采集设备真实拍摄得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，提高图像序列的检测效率，进而提高图像采集设备是否劫持的检测准确度。
本申请第三方面实施例提出了一种计算机设备，包括：存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时，实现如本申请第一方面实施例提出的图像采集设备劫持的检测方法。
本申请第四方面实施例提出了一种存储有计算机指令的非瞬时计算机可读存储介质，所述计算机指令用于使所述计算机执行本申请第一方面实施例提出的图像采集设备劫持的检测方法。
本申请第五方面实施例提出了一种计算机程序产品，当所述计算机程序产品中的指令处理器执行时，执行本申请第一方面实施例提出的图像采集设备劫持的检测方法。
本申请附加的方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本申请的实践了解到。
附图说明
本申请上述的和/或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和容易理解，其中：
图1为本申请实施例一所提供的图像采集设备劫持的检测方法的流程示意图；
图2为本申请实施例二所提供的图像采集设备劫持的检测装置的结构示意图；
图3示出了适于用来实现本申请实施方式的示例性计算机设备的框图。
具体实施方式
下面详细描述本申请的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，旨在用于解释本申请，而不能理解为对本申请的限制。
相关技术中，服务器可以通过针对合成图像序列的检测算法，来检测上传的图像序列是否为合成图像序列，进而确定上传的图像序列是否为图像采集设备实际拍摄的图像序列。但上述方法中，由于攻击方采用的图像序列合成算法多种多样，合成图像序列的检测算法难以通用于各种图像序列合成算法，导致合成图像序列检测的准确度差，进而图像采集设备是否劫持的检测准确度差。
因此，本申请主要针对现有技术中合成图像序列检测的准确度差，图像采集设备是否劫持的检测准确度差的技术问题，提出一种图像采集设备劫持的检测方法。
本申请实施例的图像采集设备劫持的检测方法，通过向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作；在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调整，并将动态调整后的拍摄参数发送给图像采集设备，以使图像采集设备采用动态调整后的拍摄参数拍摄图像；获取图像采集设备拍摄得到的图像序列；根据图像序列，生成图像采集设备的劫持检测结果，从而能够使得图像采集设备真实拍摄得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，提高图像序列的检测效率，进而提高图像采集设备是否劫持的检测准确度。
下面参考附图描述本申请实施例的图像采集设备劫持的检测方法、装置及计算机设备。
图1为本申请实施例一所提供的图像采集设备劫持的检测方法的流程示意图。
本申请实施例以该图像采集设备劫持的检测方法被配置于图像采集设备劫持的检测装置中来举例说明，该图像采集设备劫持的检测装置可以应用于任一计算机设备中，以使该计算机设备可以执行图像采集设备劫持的检测功能。
其中，计算机设备可以为个人电脑(Personal Computer，简称PC)、云端设备、移动设备等，移动设备例如可以为手机、平板电脑、个人数字助理、穿戴式设备、车载设备等具有各种操作系统、触摸屏和/或显示屏的硬件设备。
如图1所示，该图像采集设备劫持的检测方法可以包括以下步骤S101至步骤S104。
步骤101，向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作。
本申请实施例中，图像采集设备劫持的检测装置向图像采集设备发送拍摄指令的触发条件例如可以为，图像采集设备用户在登录银行账户或者转账，需要进行人脸活体检测；又例如可以为，需要检测图像采集设备是否劫持。
步骤102，在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调 整，以使图像采集设备采用动态调整后的拍摄参数进行拍摄。
本申请实施例中，图像采集设备的拍摄过程例如可以为，采用原始拍摄参数拍摄得到第一图像；接收到拍摄参数动态调整指令，对原始拍摄参数进行调整，得到调整后的拍摄参数，采用调整后的拍摄参数拍摄得到第二图像，重复上述过程，得到图像序列。
本申请实施例中，图像采集设备的拍摄参数可以包括以下参数中的至少一种：曝光时间、曝光补偿、曝光增益、焦距、分辨率、快门参数、光圈参数、是否开闪光灯等。
本申请实施例中，图像采集设备劫持的检测装置对图像采集设备拍摄参数的动态调整策略，可以为预设设置的调整策略，也可以为随机调整，可以根据实际需要进行设置。
步骤103，获取图像采集设备拍摄得到的图像序列。
步骤104，根据图像序列，生成图像采集设备的劫持检测结果。
本申请实施例中，图像采集设备劫持的检测装置执行步骤104的过程具体可以为，针对图像序列的每个相邻图像对，将相邻图像对输入至劫持检测模型，以生成相邻图像对中两个图像之间的距离；根据每个相邻图像对对应的距离，确定劫持检测结果。
本申请实施例中，劫持检测模型包括：依次连接的降维处理层和距离检测层；其中，降维处理层，用于对相邻图像对中的两个图像进行降维处理，使得图像采集设备未被劫持时降维处理后的两个图像之间的距离尽可能的小，图像采集设备被劫持时降维处理后的两个图像之间的距离尽可能的大，该点通过对劫持检测模型的训练实现；距离检测层，用于检测降维处理后的两个图像之间的距离。其中，此处的距离可以指两个图像之间的欧式距离，欧式距离可以表征两个图像之间的相似度。其中，降维处理后的图像具体可以采用向量表示，向量的维度例如可以为128维等。
本申请实施例中，为了使得图像采集设备未被劫持时降维处理后的两个图像之间的距离尽可能的小，图像采集设备被劫持时降维处理后的两个图像之间的距离尽可能的大，劫持检测模型的训练过程例如可以为，获取初始的劫持检测模型；获取训练数据，其中，训练数据包括：第一拍摄参数下的第一样本图像、图像采集设备未被劫持时第二拍摄参数下的第二样本图像以及图像采集设备被劫持时第二拍摄参数下的第三样本图像；以训练数据为输入，以第一样本图像和第二样本图像之间的第一距离，以及第一样本图像和第三样本图像之间的第二距离为输出，以第一距离和第二距离的差值构建目标函数，根据目标函数值对初始的劫持检测模型的系数进行调整，以实现训练。
其中，目标函数具体可以如以下公式(1)所示。
Figure PCTCN2022085189-appb-000001
其中，W和b表示劫持检测模型的系数，
Figure PCTCN2022085189-appb-000002
表示第一样本图像，
Figure PCTCN2022085189-appb-000003
表示第二样本图像，
Figure PCTCN2022085189-appb-000004
表示第三样本图像，f表示降维算法，
Figure PCTCN2022085189-appb-000005
表示第一距离，
Figure PCTCN2022085189-appb-000006
表示第二距离，a表示第一距离和第二距离之间的最小目标间隔；“+”号表示方括号内的值大于零时，目标函数值为方括号内的值；方括号内的值小于等于零时，目标函数值为零值。
本申请实施中，采用上述目标函数，可以使得劫持检测模型的输出中，第二距离与第一距离的差值无限靠近a值，也就说，图像采集设备未被劫持时降维处理后的两个图像之间的距离尽可能的小，图像采集设备被劫持时降维处理后的两个图像之间的距离尽可能的大，从而能够通过距离判断，来确定图像序列是合成图像序列还是真实拍摄得到的图像序列，进而确定图像采集设备是否劫持。
本申请实施例中，图像采集设备劫持的检测装置根据每个相邻图像对对应的距离，确定劫持检测结果的过程例如可以为，在存在对应的距离大于或等于预设距离阈值的相邻图像对时，确定所述图像采集设备被劫持；或者，在图像序列中对应的距离大于或等于预设距离阈值的相邻图像对的占比大于或等于预设占比阈值时，确定图像采集设备被劫持。
其中，在图像序列中对应的距离大于或等于预设距离阈值的相邻图像对的占比大于或等于预设占比阈值时，确定图像采集设备被劫持，是考虑到图像采集设备拍摄图像序列的过程中，根据拍摄对象所在位置的变化以及所在位置的光线变化等调整拍摄参数，在图像序列中对应的距离大于或等于预设距离阈值的相邻图像对的占比大于或等于预设占比阈值时，确定图像采集设备被劫持，能够进一步提高图像采集设备是否劫持的检测准确度。
本申请实施例中，在步骤104之后，所述的方法还可以包括以下步骤：在劫持检测结果为图像采集设备被劫持时，不对图像序列进行人脸识别处理；在劫持检测结果为图像采集设备未被劫持时，对图像序列进行人脸识别处理，以进行人脸活体检测等。
本申请实施例的图像采集设备劫持的检测方法，通过向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作；在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调整，并将动态调整后的拍摄参数发送给图像采集设备，以使图像采集设备采用动态调整后的拍摄参数拍摄图像；获取图像采集设备拍摄得到的图像序列；根据图像序列，生成图像采集设备的劫持检测结果，从而能够使得图像采集设备真实拍摄得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，提高图像序列的检测效率，进而提高图像采集设备是否劫持的检测准确度。
图2为本申请实施例二所提供的图像采集设备劫持的检测装置的结构示意图。
如图2所示，该图像采集设备劫持的检测装置200可以包括：发送模块210、动态调整模块220、获取模块230和生成模块240。
其中，发送模块210，用于向图像采集设备发送拍摄指令，以使所述图像采集设备进行拍摄操作；
动态调整模块220，用于在所述图像采集设备拍摄的过程中，对所述图像采集设备的拍摄参数进行动态调整，以使所述图像采集设备采用动态调整后的拍摄参数进行拍摄；
获取模块230，用于获取所述图像采集设备拍摄得到的图像序列；
生成模块240，用于根据所述图像序列，生成所述图像采集设备的劫持检测结果。
进一步地，在本申请实施例中，所述生成模块240具体用于，
针对所述图像序列的每个相邻图像对，将所述相邻图像对输入至劫持检测模型，以生成所述相邻图像对中两个图像之间的距离；
根据每个相邻图像对对应的距离，确定劫持检测结果。
进一步地，在本申请实施例中，所述劫持检测模型包括：依次连接的降维处理层和距离检测层；
其中，所述降维处理层，用于对所述相邻图像对中的两个图像进行降维处理；
所述距离检测层，用于检测降维处理后的所述两个图像之间的距离。
进一步地，在本申请实施例中，所述的装置还包括：训练模块；
所述获取模块230，还用于获取初始的劫持检测模型；
所述获取模块230，还用于获取训练数据，其中，所述训练数据包括：第一拍摄参数下的第一样本图像、图像采集设备未被劫持时第二拍摄参数下的第二样本图像以及图像采集设备被劫持时第二拍摄参数下的第三样本图像；
所述训练模块，用于以所述训练数据为输入，以所述第一样本图像和所述第二样本图像之间的第一距离，以及所述第一样本图像和所述第三样本图像之间的第二距离为输出，以所述第一距离和所述第二距离的差值构建目标函数，根据目标函数值对初始的劫持检测模型的系数进行调整，以实现训练。
进一步地，在本申请实施例中，所述目标函数为，
Figure PCTCN2022085189-appb-000007
其中，W和b表示劫持检测模型的系数，
Figure PCTCN2022085189-appb-000008
表示第一样本图像，
Figure PCTCN2022085189-appb-000009
表示第二样本图像，
Figure PCTCN2022085189-appb-000010
表示第三样本图像，f表示降维算法，
Figure PCTCN2022085189-appb-000011
表示第一距离，
Figure PCTCN2022085189-appb-000012
表示第二距离，a表示第一距离和第二距离之间的最小目标间隔。
进一步地，在本申请实施例中，所述生成模块240具体用于，在存在对应的距离大于或等于预设距离阈值的相邻图像对时，确定所述图像采集设备被劫持；或者，在所述 图像序列中对应的距离大于或等于预设距离阈值的相邻图像对的占比大于或等于预设占比阈值时，确定所述图像采集设备被劫持。
进一步地，在本申请实施例中，所述动态调整模块具体用于，对所述图像采集设备的拍摄参数进行随机动态调整。
进一步地，在本申请实施例中，所述的装置还包括：处理模块，用于在所述劫持检测结果为所述图像采集设备被劫持时，不对所述图像序列进行人脸识别处理；在所述劫持检测结果为所述图像采集设备未被劫持时，对所述图像序列进行人脸识别处理。
需要说明的是，前述实施例一中的解释说明也适用于该实施例的图像采集设备劫持的检测装置，此处不再赘述。
本申请实施例的图像采集设备劫持的检测装置，通过向图像采集设备发送拍摄指令，以使图像采集设备进行拍摄操作；在图像采集设备拍摄的过程中，对图像采集设备的拍摄参数进行动态调整，并将动态调整后的拍摄参数发送给图像采集设备，以使图像采集设备采用动态调整后的拍摄参数拍摄图像；获取图像采集设备拍摄得到的图像序列；根据图像序列，生成图像采集设备的劫持检测结果，从而能够使得图像采集设备真实拍摄得到的图像序列中不同的拍摄参数对应不同图像，而图像采集设备劫持中不同的拍摄参数对应不同图像时会出现异常，从而能够区分合成图像序列以及真实拍摄得到的图像序列，提高图像序列的检测效率，进而提高图像采集设备是否劫持的检测准确度。
为了实现上述实施例，本申请还可以提出一种图像采集设备劫持的检测系统，包括：终端设备和云端设备；其中，终端设备上设置有图像采集设备，云端设备可以与终端设备连接，执行如上实施例中所述的图像采集设备劫持的检测方法。其中，云端设备上可以设置有检测服务，或者调用其他设备上的检测服务，根据图像序列生成图像采集设备的劫持检测结果。
为了实现上述实施例，本申请还提出一种计算机设备，包括：存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，处理器执行程序时，实现如本申请前述实施例提出的图像采集设备劫持的检测方法。
为了实现上述实施例，本申请还提出一种计算机程序产品，当计算机程序产品中的指令处理器执行时，执行如本申请前述实施例提出的图像采集设备劫持的检测方法。
图3示出了适于用来实现本申请实施方式的示例性计算机设备的框图。图3显示的计算机设备12仅仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。
如图3所示，计算机设备12以通用计算设备的形式表现。计算机设备12的组件可以包括但不限于：一个或者多个处理器或者处理单元16，系统存储器28，连接不同系统组件(包括系统存储器28和处理单元16)的总线18。
总线18表示几类总线结构中的一种或多种，包括存储器总线或者存储器控制器，外围总线，图形加速端口，处理器或者使用多种总线结构中的任意总线结构的局域总 线。举例来说，这些体系结构包括但不限于工业标准体系结构(Industry Standard Architecture；以下简称：ISA)总线，微通道体系结构(Micro Channel Architecture；以下简称：MAC)总线，增强型ISA总线、视频电子标准协会(Video Electronics Standards Association；以下简称：VESA)局域总线以及外围组件互连(Peripheral Component Interconnection；以下简称：PCI)总线。
计算机设备12典型地包括多种计算机系统可读介质。这些介质可以是任何能够被计算机设备12访问的可用介质，包括易失性和非易失性介质，可移动的和不可移动的介质。
存储器28可以包括易失性存储器形式的计算机系统可读介质，例如随机存取存储器(Random Access Memory；以下简称：RAM)30和/或高速缓存存储器32。计算机设备12可以进一步包括其它可移动/不可移动的、易失性/非易失性计算机系统存储介质。仅作为举例，存储系统34可以用于读写不可移动的、非易失性磁介质(图3未显示，通常称为“硬盘驱动器”)。尽管图3中未示出，可以提供用于对可移动非易失性磁盘(例如“软盘”)读写的磁盘驱动器，以及对可移动非易失性光盘(例如：光盘只读存储器(Compact Disc Read Only Memory；以下简称：CD-ROM)、数字多功能只读光盘(Digital Video Disc Read Only Memory；以下简称：DVD-ROM)或者其它光介质)读写的光盘驱动器。在这些情况下，每个驱动器可以通过一个或者多个数据介质接口与总线18相连。存储器28可以包括至少一个程序产品，该程序产品具有一组(例如至少一个)程序模块，这些程序模块被配置以执行本申请各实施例的功能。
具有一组(至少一个)程序模块42的程序/实用工具40，可以存储在例如存储器28中，这样的程序模块42包括但不限于操作系统、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可以包括网络环境的实现。程序模块42通常执行本申请所描述的实施例中的功能和/或方法。
计算机设备12也可以与一个或多个外部设备14(例如键盘、指向设备、显示器24等)通信，还可与一个或者多个使得用户能与该计算机设备12交互的设备通信，和/或与使得该计算机设备12能与一个或多个其它计算设备进行通信的任何设备(例如网卡，调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口22进行。并且，计算机设备12还可以通过网络适配器20与一个或者多个网络(例如局域网(Local Area Network；以下简称：LAN)，广域网(Wide Area Network；以下简称：WAN)和/或公共网络，例如因特网)通信。如图所示，网络适配器20通过总线18与计算机设备12的其它模块通信。应当明白，尽管图中未示出，可以结合计算机设备12使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。
处理单元16通过运行存储在系统存储器28中的程序，从而执行各种功能应用以及数据处理，例如实现前述实施例中提及的方法。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示 例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本申请的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现定制逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本申请的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本申请的实施例所属技术领域的技术人员所理解。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，"计算机可读介质"可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本申请的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。如，如果用硬件来实现和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储 介质中，该程序在执行时，包括方法实施例的步骤之一或其组合。
此外，在本申请各个实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。
上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本申请的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本申请的限制，本领域的普通技术人员在本申请的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (19)

  1. 一种图像采集设备劫持的检测方法，包括：
    向图像采集设备发送拍摄指令，以使所述图像采集设备进行拍摄操作；
    在所述图像采集设备拍摄的过程中，对所述图像采集设备的拍摄参数进行动态调整，以使所述图像采集设备采用动态调整后的拍摄参数进行拍摄；
    获取所述图像采集设备拍摄得到的图像序列；
    根据所述图像序列，生成所述图像采集设备的劫持检测结果。
  2. 根据权利要求1所述的方法，其中，所述根据所述图像序列，生成所述图像采集设备的劫持检测结果，包括：
    针对所述图像序列的每个相邻图像对，将所述相邻图像对输入至劫持检测模型，以生成所述相邻图像对中两个图像之间的距离；
    根据每个相邻图像对对应的距离，确定劫持检测结果。
  3. 根据权利要求2所述的方法，其中，所述劫持检测模型包括：依次连接的降维处理层和距离检测层；
    其中，所述降维处理层，用于对所述相邻图像对中的两个图像进行降维处理；
    所述距离检测层，用于检测降维处理后的所述两个图像之间的距离。
  4. 根据权利要求3所述的方法，其中，在将所述相邻图像对输入至劫持检测模型之前，还包括：
    获取初始的劫持检测模型；
    获取训练数据，其中，所述训练数据包括：第一拍摄参数下的第一样本图像、图像采集设备未被劫持时第二拍摄参数下的第二样本图像以及图像采集设备被劫持时第二拍摄参数下的第三样本图像；
    以所述训练数据为输入，以所述第一样本图像和所述第二样本图像之间的第一距离，以及所述第一样本图像和所述第三样本图像之间的第二距离为输出，以所述第一距离和所述第二距离的差值构建目标函数，根据目标函数值对初始的劫持检测模型的系数进行调整，以实现训练。
  5. 根据权利要求4所述的方法，其中，所述目标函数为，
    Figure PCTCN2022085189-appb-100001
    其中，W和b表示劫持检测模型的系数，
    Figure PCTCN2022085189-appb-100002
    表示第一样本图像，
    Figure PCTCN2022085189-appb-100003
    表示第二样本图像，
    Figure PCTCN2022085189-appb-100004
    表示第三样本图像，f表示降维算法，
    Figure PCTCN2022085189-appb-100005
    表示第一距离，
    Figure PCTCN2022085189-appb-100006
    表示第二距离，a表示第一距离和第二距离之间的最小目标间隔。
  6. 根据权利要求2所述的方法，其中，所述根据每个相邻图像对对应的距离，确定劫持检测结果，包括：
    在存在对应的距离大于或等于预设距离阈值的相邻图像对时，确定所述图像采集设备被劫持；
    或者，
    在所述图像序列中对应的距离大于或等于预设距离阈值的相邻图像对的占比大于或等于预设占比阈值时，确定所述图像采集设备被劫持。
  7. 根据权利要求1所述的方法，其中，所述对所述图像采集设备的拍摄参数进行动态调整，包括：
    对所述图像采集设备的拍摄参数进行随机动态调整。
  8. 根据权利要求1所述的方法，还包括：
    在所述劫持检测结果为所述图像采集设备被劫持时，不对所述图像序列进行人脸识别处理；
    在所述劫持检测结果为所述图像采集设备未被劫持时，对所述图像序列进行人脸识别处理。
  9. 一种图像采集设备劫持的检测装置，其中，包括：
    发送模块，用于向图像采集设备发送拍摄指令，以使所述图像采集设备进行拍摄操作；
    动态调整模块，用于在所述图像采集设备拍摄的过程中，对所述图像采集设备的拍摄参数进行动态调整，以使所述图像采集设备采用动态调整后的拍摄参数进行拍摄；
    获取模块，用于获取所述图像采集设备拍摄得到的图像序列；
    生成模块，用于根据所述图像序列，生成所述图像采集设备的劫持检测结果。
  10. 根据权利要求9所述的装置，其中，所述生成模块具体用于，
    针对所述图像序列的每个相邻图像对，将所述相邻图像对输入至劫持检测模型，以生成所述相邻图像对中两个图像之间的距离；
    根据每个相邻图像对对应的距离，确定劫持检测结果。
  11. 根据权利要求10所述的装置，其中，所述劫持检测模型包括：依次连接的降维处理层和距离检测层；
    其中，所述降维处理层，用于对所述相邻图像对中的两个图像进行降维处理；
    所述距离检测层，用于检测降维处理后的所述两个图像之间的距离。
  12. 根据权利要求11所述的装置，还包括：训练模块；
    所述获取模块，还用于获取初始的劫持检测模型；
    所述获取模块，还用于获取训练数据，其中，所述训练数据包括：第一拍摄参数下的第一样本图像、图像采集设备未被劫持时第二拍摄参数下的第二样本图像以及图像采集设备被劫持时第二拍摄参数下的第三样本图像；
    所述训练模块，用于以所述训练数据为输入，以所述第一样本图像和所述第二样本图像之间的第一距离，以及所述第一样本图像和所述第三样本图像之间的第二距离为输出，以所述第一距离和所述第二距离的差值构建目标函数，根据目标函数值对初 始的劫持检测模型的系数进行调整，以实现训练。
  13. 根据权利要求12所述的装置，其中，所述目标函数为，
    Figure PCTCN2022085189-appb-100007
    其中，W和b表示劫持检测模型的系数，
    Figure PCTCN2022085189-appb-100008
    表示第一样本图像，
    Figure PCTCN2022085189-appb-100009
    表示第二样本图像，
    Figure PCTCN2022085189-appb-100010
    表示第三样本图像，f表示降维算法，
    Figure PCTCN2022085189-appb-100011
    表示第一距离，
    Figure PCTCN2022085189-appb-100012
    表示第二距离，a表示第一距离和第二距离之间的最小目标间隔。
  14. 根据权利要求10所述的装置，其中，所述生成模块具体用于，
    在存在对应的距离大于或等于预设距离阈值的相邻图像对时，确定所述图像采集设备被劫持；
    或者，
    在所述图像序列中对应的距离大于或等于预设距离阈值的相邻图像对的占比大于或等于预设占比阈值时，确定所述图像采集设备被劫持。
  15. 根据权利要求9所述的装置，其中，所述动态调整模块具体用于，
    对所述图像采集设备的拍摄参数进行随机动态调整。
  16. 根据权利要求9所述的装置，还包括：
    处理模块，用于在所述劫持检测结果为所述图像采集设备被劫持时，不对所述图像序列进行人脸识别处理；
    在所述劫持检测结果为所述图像采集设备未被劫持时，对所述图像序列进行人脸识别处理。
  17. 一种电子设备，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行权利要求1至8中任一项所述的方法。
  18. 一种存储有计算机指令的非瞬时计算机可读存储介质，其中，所述计算机指令用于使所述计算机执行权利要求1至8中任一项所述的方法。
  19. 一种计算机程序产品，包括计算机程序，当所述计算机程序在被处理器执行时，实现根据权利要求1至8中任一项所述的方法。
PCT/CN2022/085189 2021-04-06 2022-04-02 图像采集设备劫持的检测方法、装置及计算机设备 WO2022213955A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110369263.XA CN113807160B (zh) 2021-04-06 2021-04-06 图像采集设备劫持的检测方法、装置及计算机设备
CN202110369263.X 2021-04-06

Publications (1)

Publication Number Publication Date
WO2022213955A1 true WO2022213955A1 (zh) 2022-10-13

Family

ID=78892979

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/085189 WO2022213955A1 (zh) 2021-04-06 2022-04-02 图像采集设备劫持的检测方法、装置及计算机设备

Country Status (2)

Country Link
CN (1) CN113807160B (zh)
WO (1) WO2022213955A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113807160B (zh) * 2021-04-06 2024-02-06 京东科技控股股份有限公司 图像采集设备劫持的检测方法、装置及计算机设备
CN115174138B (zh) * 2022-05-25 2024-06-07 北京旷视科技有限公司 摄像头攻击检测方法、系统、设备、存储介质及程序产品
CN116486140B (zh) * 2023-03-22 2024-04-05 中化现代农业有限公司 土壤质地分类方法、装置和电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111091112A (zh) * 2019-12-30 2020-05-01 支付宝实验室(新加坡)有限公司 活体检测方法及装置
CN111144365A (zh) * 2019-12-31 2020-05-12 北京三快在线科技有限公司 活体检测方法、装置、计算机设备及存储介质
CN112395902A (zh) * 2019-08-12 2021-02-23 北京旷视科技有限公司 人脸活体检测方法、图像分类方法、装置、设备和介质
CN113807160A (zh) * 2021-04-06 2021-12-17 京东科技控股股份有限公司 图像采集设备劫持的检测方法、装置及计算机设备

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012073233A1 (en) * 2010-11-29 2012-06-07 Biocatch Ltd. Method and device for confirming computer end-user identity
WO2014039763A1 (en) * 2012-09-09 2014-03-13 Michael Fiske Visual image authentication and transaction authorization using non-determinism
CN107609462A (zh) * 2017-07-20 2018-01-19 北京百度网讯科技有限公司 待检测信息生成及活体检测方法、装置、设备及存储介质
KR102037419B1 (ko) * 2017-09-05 2019-10-28 삼성전자주식회사 영상 표시 장치 및 그 동작 방법
CN110633659B (zh) * 2019-08-30 2022-11-04 北京旷视科技有限公司 活体检测方法、装置、计算机设备和存储介质
CN110807368B (zh) * 2019-10-08 2022-04-29 支付宝(杭州)信息技术有限公司 一种注入攻击的识别方法、装置及设备
CN111881844B (zh) * 2020-07-30 2021-05-07 北京嘀嘀无限科技发展有限公司 一种判断图像真实性的方法及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112395902A (zh) * 2019-08-12 2021-02-23 北京旷视科技有限公司 人脸活体检测方法、图像分类方法、装置、设备和介质
CN111091112A (zh) * 2019-12-30 2020-05-01 支付宝实验室(新加坡)有限公司 活体检测方法及装置
CN111144365A (zh) * 2019-12-31 2020-05-12 北京三快在线科技有限公司 活体检测方法、装置、计算机设备及存储介质
CN113807160A (zh) * 2021-04-06 2021-12-17 京东科技控股股份有限公司 图像采集设备劫持的检测方法、装置及计算机设备

Also Published As

Publication number Publication date
CN113807160B (zh) 2024-02-06
CN113807160A (zh) 2021-12-17

Similar Documents

Publication Publication Date Title
WO2022213955A1 (zh) 图像采集设备劫持的检测方法、装置及计算机设备
US20210279469A1 (en) Image signal provenance attestation
US10237473B2 (en) Depth map calculation in a stereo camera system
US10438322B2 (en) Image resolution enhancement
WO2020018359A1 (en) Three-dimensional living-body face detection method, face authentication recognition method, and apparatuses
US11080553B2 (en) Image search method and apparatus
US20180342045A1 (en) Image resolution enhancement using machine learning
JP5199471B2 (ja) 色恒常性方法及びシステム
JP4781233B2 (ja) 画像処理装置、撮像装置、及び画像処理方法
JP5536010B2 (ja) 電子カメラ、撮像制御プログラム及び撮像制御方法
CN111091590A (zh) 图像处理方法、装置、存储介质及电子设备
WO2006033797A1 (en) Method and apparatus for automatic image orientation normalization
KR102344041B1 (ko) 병변 진단 시스템 및 방법
WO2018058476A1 (zh) 一种图像校正方法及装置
JP7204786B2 (ja) 視覚的検索方法、装置、コンピュータ機器及び記憶媒体
JP6932758B2 (ja) 物体検出装置、物体検出方法、物体検出プログラム、学習装置、学習方法及び学習プログラム
KR102037997B1 (ko) 전자 장치 및 콘텐츠 생성 방법
CN111126101B (zh) 关键点位置的确定方法、装置、电子设备和存储介质
US11823359B2 (en) Systems and methods for leveling images
JP6933125B2 (ja) 情報処理装置，撮影ガイド表示プログラム，撮影ガイド表示方法
JP2009042900A (ja) 撮像装置および画像選択装置
TWM583989U (zh) 序號檢測系統
CN114827473B (zh) 视频处理方法和装置
TWI703504B (zh) 序號檢測系統
WO2023100332A1 (ja) 画像生成装置、画像生成方法および記録媒体

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22784021

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 20.02.2024)

122 Ep: pct application non-entry in european phase

Ref document number: 22784021

Country of ref document: EP

Kind code of ref document: A1