WO2023178997A1 - 交互控制方法、装置、计算机设备及存储介质 - Google Patents

交互控制方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2023178997A1
WO2023178997A1 PCT/CN2022/125855 CN2022125855W WO2023178997A1 WO 2023178997 A1 WO2023178997 A1 WO 2023178997A1 CN 2022125855 W CN2022125855 W CN 2022125855W WO 2023178997 A1 WO2023178997 A1 WO 2023178997A1
Authority
WO
WIPO (PCT)
Prior art keywords
color
face image
target
face
detection
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2022/125855
Other languages
English (en)
French (fr)
Inventor
张殿炎
旷章辉
于博文
郝石磊
强丽丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Sensetime Intelligent Technology Co Ltd
Original Assignee
Shanghai Sensetime Intelligent Technology Co Ltd
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 Shanghai Sensetime Intelligent Technology Co Ltd filed Critical Shanghai Sensetime Intelligent Technology Co Ltd
Publication of WO2023178997A1 publication Critical patent/WO2023178997A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
    • Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection

Definitions

  • the present disclosure relates to the field of identity recognition technology, and specifically relates to an interactive control method, device, computer equipment and storage medium.
  • Embodiments of the present disclosure provide at least an interactive control method, device, computer equipment, and storage medium.
  • embodiments of the present disclosure provide an interactive control method, which method includes: in response to obtaining a target interaction request, obtaining a first color sequence; the first color sequence includes: multiple colors; based on the first color Sequence, obtain the first face image corresponding to the target face in multiple colors in the first color sequence; based on the first color sequence, perform color detection and liveness detection on the first face image , and perform identity detection based on the first face image; in response to the first face image passing the color detection, the living body detection, and the identity detection, performing an interaction corresponding to the target interaction request operate.
  • the recognition accuracy and recognition speed can improve the efficiency of verifying user information and the accuracy of identifying users in driving schools, prevent tutoring behavior, and reduce safety hazards caused by tutoring behavior.
  • obtaining the first face images corresponding to the target face in multiple colors in the first color sequence based on the first color sequence includes: based on the first color sequence Color sequence, control the target device to sequentially emit light corresponding to multiple colors in the first color sequence, and when controlling the target device to emit light of each color, obtain the first light corresponding to the color. Face images.
  • the first face image can be acquired in the light corresponding to the first color sequence, thereby reducing the possibility of being counterfeited during the acquisition process of the first face image, and improving the security of face detection.
  • controlling the target device to sequentially emit light corresponding to multiple colors in the first color sequence includes: detecting whether the target face is located behind the light emitted by the target device. within the light coverage area; in response to the target face being located within the light coverage area, controlling the target device to sequentially emit light corresponding to multiple colors in the first color sequence.
  • performing color detection on the first face image based on the first color sequence includes: for each color in the first color sequence, for the third color corresponding to the color Perform color detection on a face image to obtain the color information of the first face image corresponding to the color; match each color with the color information of the corresponding first face image; respond to the first color sequence
  • the multiple colors in are successfully matched with the color information corresponding to the first face image, and it is determined that the first face image passes the color detection.
  • performing liveness detection on the first face image includes: for each color in the first color sequence, from the first face image corresponding to each color , determine the first target face image corresponding to each color; perform live detection on the first target face image corresponding to each color, and obtain the first target face image corresponding to each color.
  • the living body detection result in response to the live body detection result indicating that the number of first target face images of which the target face is a live human face reaches a preset number, it is determined that the first face image passes the live body detection.
  • the liveness detection when the liveness detection is performed on the first face image, it will be determined that the first face image passes the liveness detection only when the number of first target face images whose liveness detection results indicate that the target face is a live face reaches the preset number. , thereby ensuring the accuracy of in vivo detection results.
  • determining the first target face image corresponding to each color from the first face image corresponding to each color includes: determining the first face image corresponding to each color. quality information of the first face image; based on the quality information, determine the first target face image corresponding to each color.
  • the first target face image corresponding to each color is determined through the quality information of the first face image corresponding to each color, so that the determined first target face image has a higher accuracy when performing life detection. Detection accuracy.
  • the living body detection includes: in response to the live body detection result indicating that the target face is a living human face, the number of first target face images reaches a preset number, and the live body detection result indicates that the target face is a living human face.
  • the quality information corresponding to the first target face image meets the preset conditions, and it is determined that the first face image passes the living body detection.
  • the live body detection result indicates the quality corresponding to the first target face image in which the target face is a living face. Whether the information meets the preset conditions can more accurately determine whether the first face image can pass the liveness detection.
  • the identity detection based on the first face image includes: based on the quality information of the first face image corresponding to each color in the first color sequence, determining the second face image.
  • Target face image perform identity detection based on the second target face image and the pre-stored face registration image.
  • the method further includes: in response to receiving the face registration request, obtaining a second color sequence; the second color sequence includes: multiple colors; based on the second color sequence, obtaining the target face in Second face images corresponding to multiple colors in the second color sequence; performing color detection on the second face image based on the second color sequence, and performing color detection on the second face image
  • the image is subjected to live body detection; in response to the second face image passing the color detection and the live body detection, the face registration image is determined based on the second face image.
  • the face registration image is used to determine the face registration image, so that when performing identity detection on the first face image, the face registration image is provided, thereby achieving higher recognition accuracy and recognition speed during the identity detection process, and increasing reliability. sex.
  • the target interaction request includes at least one of the following: a login request, a course purchase request, a reservation request, a class registration request, and a vehicle practice operation request.
  • the target interaction request includes: a reservation request, and performing an interactive operation corresponding to the target interaction request includes:
  • course reservation information Obtain course reservation information, make course reservations based on the course reservation information, and generate course reservation records.
  • the target interaction request includes: a class registration request
  • performing the interactive operation corresponding to the target interaction request includes: obtaining a course reservation record; indicating a reservation in response to the course reservation record
  • the course start time meets the preset time conditions, and the security inspection equipment is controlled to open the release channel.
  • the target interaction request includes: a vehicle practical operation request
  • performing the interactive operation corresponding to the target interaction request includes: obtaining a course registration record; based on the course registration record, controlling The user information corresponding to the course registration record enters the queuing state; in response to the user information entering the queuing state reaching the queuing end condition, the user information is assigned a target vehicle for vehicle operation practice.
  • the user information corresponding to the course registration record is controlled to enter the queuing state.
  • the user information entering the queuing state reaches the queuing end condition, the user can perform vehicle operation practice, ensuring that the user who performs vehicle operation practice and The user registered for the course must be the same person to prevent substitute teaching behavior.
  • embodiments of the present disclosure also provide an interactive control device, including:
  • An acquisition module configured to acquire a first color sequence in response to acquiring a target interaction request; the first color sequence includes: multiple colors; based on the first color sequence, acquire the target face in the first color sequence Corresponding first face images in multiple colors;
  • a detection module configured to perform color detection and live body detection on the first face image based on the first color sequence, and perform identity detection based on the first face image;
  • a processing module configured to perform an interactive operation corresponding to the target interaction request in response to the first face image passing the color detection, the live body detection, and the identity detection.
  • the acquisition module when the acquisition module acquires the first face images corresponding to the target face in multiple colors in the first color sequence based on the first color sequence, For controlling the target device to sequentially emit light corresponding to multiple colors in the first color sequence based on the first color sequence, and when controlling the target device to emit light of each color, obtain and The first face image corresponding to this color.
  • the detection module is configured to detect whether the target face is located when the control target device sequentially emits light corresponding to multiple colors in the first color sequence. Within the light coverage area after the target device emits light; in response to the target face being located within the light coverage area, control the target device to sequentially emit light corresponding to multiple colors in the first color sequence.
  • the detection module when performing color detection on the first face image based on the first color sequence, is configured to detect each color in the first color sequence. , perform color detection on the first face image corresponding to the color, and obtain the color information of the first face image corresponding to the color; match each color with the color information of the corresponding first face image; respond If the multiple colors in the first color sequence are successfully matched with the color information corresponding to the first face image, it is determined that the first face image passes the color detection.
  • the detection module when performing liveness detection on the first face image, is configured to detect, for each color in the first color sequence, from each color of the first face image. In the first face image corresponding to the color, determine the first target face image corresponding to each color; perform in vivo detection on the first target face image corresponding to each color to obtain the corresponding color The liveness detection result of the first target face image; in response to the liveness detection result indicating that the number of first target face images of the target face being a live face reaches a preset number, it is determined that the first face image passes The living body detection.
  • the method further includes a determining module, configured to determine the first target face image corresponding to each color from the first face image corresponding to each color. Determine the quality information of the first face image corresponding to each color; determine the first target face image corresponding to each color based on the quality information.
  • a determining module configured to determine the first target face image corresponding to each color from the first face image corresponding to each color. Determine the quality information of the first face image corresponding to each color; determine the first target face image corresponding to each color based on the quality information.
  • the determination module determines the first target face image when the number of first target face images indicating that the target face is a live face in response to a live body detection result reaches a preset number.
  • the number of first target face images used to indicate that the target face is a live face in response to the live body detection result reaches a preset number, and the live body detection result indicates that the target If the quality information corresponding to the first target face image whose face is a living face meets the preset conditions, it is determined that the first face image passes the living body detection.
  • the detection module when performing identity detection based on the first face image, is used to detect the identity based on the first face image corresponding to each color in the first color sequence. quality information, determine the second target face image; perform identity detection based on the second target face image and the pre-stored face registration image.
  • the detection module is configured to obtain a second color sequence in response to receiving a face registration request; the second color sequence includes: multiple colors; based on the second color sequence, obtain A second face image corresponding to the target face in multiple colors in the second color sequence; performing color detection on the second face image based on the second color sequence, and performing color detection on the second face image.
  • the second face image is subjected to live body detection; in response to the second face image passing the color detection and the live body detection, the face registration image is determined based on the second face image.
  • the target interaction request includes at least one of the following:
  • the target interaction request includes: a reservation request
  • the processing module is configured to obtain course reservation information when performing an interaction operation corresponding to the target interaction request, and based on the Use the above course reservation information to make course reservations and generate course reservation records.
  • the target interaction request includes: a class registration request, and the processing module is used to obtain course reservation records when performing an interaction operation corresponding to the target interaction request; in response to The course reservation record indicates that the scheduled start time of the course meets the preset time conditions, and the security inspection equipment is controlled to open the release channel.
  • the target interaction request includes: a vehicle practice operation request, and the processing module is used to obtain course registration records when performing the interaction operation corresponding to the target interaction request; based on The course registration record controls the user information corresponding to the course registration record to enter the queuing state; in response to the user information entering the queuing state reaching the queuing end condition, the user information is assigned a target vehicle for vehicle operation practice.
  • an optional implementation manner of the present disclosure also provides a computer device, a processor, and a memory.
  • the memory stores machine-readable instructions executable by the processor, and the processor is configured to execute the instructions stored in the memory.
  • Machine-readable instructions when the machine-readable instructions are executed by the processor, when the machine-readable instructions are executed by the processor, the above-mentioned first aspect, or any possible implementation of the first aspect, is executed. steps in the way.
  • an optional implementation manner of the present disclosure also provides a computer-readable storage medium.
  • the computer-readable storage medium stores a computer program. When the computer program is run, it executes the above-mentioned first aspect, or any of the first aspects. steps in a possible implementation.
  • Figure 1 shows a flow chart of an interactive control method provided by an embodiment of the present disclosure
  • Figure 2 shows a flow chart of a specific example of color detection on the first face image provided by an embodiment of the present disclosure
  • Figure 3 shows a flowchart of a specific example of life detection on a first face image provided by an embodiment of the present disclosure
  • Figure 4 shows a flow chart of a specific example of identity detection based on a first face image provided by an embodiment of the present disclosure
  • Figure 5 shows a flow chart of a specific method for obtaining a face registration image provided by an embodiment of the present disclosure
  • Figure 6 shows a schematic diagram of an interactive control device provided by an embodiment of the present disclosure
  • FIG. 7 shows a schematic diagram of a computer device provided by an embodiment of the present disclosure.
  • the present disclosure provides an interactive control method by acquiring a first face image under a certain first color sequence, and performing color detection, liveness detection and identity based on the acquired first face image.
  • the detection ensures the identity of the user who initiated the target interaction request, with higher recognition accuracy and recognition speed, improves the efficiency of verifying user information and the accuracy of identifying users in driving schools, prevents the occurrence of substitute lessons, and reduces the risk of substitute lessons. resulting in safety hazards.
  • the execution subject of the interactive control method provided by the embodiment of the present disclosure is generally a computer device with certain computing capabilities.
  • the computer Devices include, for example, terminal devices or servers or other processing devices.
  • the terminal devices may be computer devices, such as computers, tablets, etc.
  • the interactive control method can be implemented by the processor calling computer-readable instructions stored in the memory.
  • the interactive control method provided by the embodiments of the present disclosure can also be used in other scenarios, such as banking systems, access control systems, etc.
  • the embodiments of the present disclosure do not limit the specific application scenarios.
  • FIG. 1 a flow chart of an interactive control method provided by an embodiment of the present disclosure is shown.
  • the method includes steps S101 to S104, wherein:
  • S101 In response to obtaining the target interaction request, obtain a first color sequence; the first color sequence includes: multiple colors;
  • S103 Based on the first color sequence, perform color detection and life detection on the first face image, and perform identity detection based on the first face image;
  • S104 In response to the first face image passing the color detection, the living body detection, and the identity detection, perform an interactive operation corresponding to the target interaction request.
  • the embodiments of the present disclosure mainly obtain a first color sequence in response to obtaining a target interaction request, and based on the first color sequence, obtain the first faces corresponding to the target face in multiple colors in the first color sequence.
  • Image perform color detection and liveness detection on the first face image, and perform identity detection based on the first face image, when the first face image passes the color detection, the liveness detection, and After the identity detection, the interactive operation corresponding to the target interaction request is executed, which ensures the identity of the user who initiated the target interaction request, has higher recognition accuracy and recognition speed, and improves the efficiency of verifying user information and identification in driving schools.
  • the user's accuracy can prevent the occurrence of substitute teaching behavior and reduce the safety hazards caused by substitute teaching behavior.
  • the target interaction request includes, for example, at least one of the following: a login request, a course purchase request, a reservation request, a class registration request, and a vehicle practice operation request.
  • an application corresponding to the driving school can be deployed; through the application, the user can enter the login page and initiate the login page.
  • the terminal device may be a mobile terminal held by the user; it may also be a dedicated terminal device at a certain location.
  • a special driving school course service device is set up in the service hall of the driving school. Through the driving school course service device, users can register, log in, purchase courses and other operations by themselves.
  • a course purchase request is a request to purchase a driving school course initiated through the service page of the application.
  • the interactive control method provided by the embodiments of the present disclosure can be applied to terminal devices and servers.
  • the terminal device or server may obtain the first color sequence in response to obtaining the target interaction request,
  • the user when the interactive control method is applied to the terminal device, the user triggers the target interaction request through the terminal device; the terminal device responds to the target interaction request triggered by the user and sends a request to obtain the first color sequence to the server, and the server receives the request from the terminal
  • the request sent by the device to obtain the first color sequence generates the first color sequence based on the preset color sequence generation rules, and returns the first color sequence to the terminal device.
  • the server can receive the target interaction request initiated by the terminal device. After receiving the target interaction request initiated by the terminal device, the server directly generates the first color sequence according to the preset color sequence generation rules.
  • the generation rules of the color sequence include, for example: restrictions on the type of colors that can be selected in the color sequence, and restrictions on the number of colors in the color sequence.
  • available colors include: red, green, yellow, blue, purple, orange, pink;
  • the number of colors in the color sequence is 5.
  • the color sequence generation rules can be set according to actual needs, and are not limited in the embodiments of the present disclosure. in:
  • the colors in the color sequences obtained by different target interaction requests can be the same (in this case, the positions of each color in different color sequences are different).
  • the colors corresponding to the first colors for different target interaction requests include the same color but different color positions.
  • the first color sequence obtained for the first time is : red, yellow, green
  • the first color sequence obtained for the second time is: green, red, yellow.
  • the colors in the first color sequence obtained by different target interaction requests are completely different, when the target interaction request is obtained in response to the target interaction request, the colors in the first color sequence obtained may be different.
  • the first time in response to the acquisition After receiving the target interaction request the first color sequence obtained for the first time is red, yellow, and green; after the target interaction request is obtained for the second time, the first color sequence obtained for the second time is white, purple, and orange.
  • the colors in the color sequence obtained by different target interaction requests are not all the same, when the target interaction request is obtained in response to the target interaction request, at least one color in the first color sequence obtained may be different; for example, the first time In response to obtaining the target interaction request, the first color sequence obtained for the first time is red, yellow, and green; after responding to the target interaction request for the second time, the first color sequence obtained for the second time is red, yellow. ,blue.
  • the embodiment of the present disclosure provides a specific method of obtaining the first face images corresponding to the target face in multiple colors in the first color sequence based on the first color sequence, including :
  • control the target device Based on the first color sequence, control the target device to sequentially emit light corresponding to multiple colors in the first color sequence, and when controlling the target device to emit light of each color, obtain the light corresponding to the color.
  • the color corresponds to the first face image.
  • the target face is the user's living face that the image collection device needs to collect. It cannot be the user's photo or user's face image video to prevent fake face information through photos or videos, causing certain security risks.
  • the target device is controlled to sequentially emit light corresponding to multiple colors in the first color sequence, so that the light reflected by the human face under the multiple colors of light has a one-to-one corresponding color. Obtain first face images corresponding to multiple colors of light.
  • the control end that controls the target device to sequentially emit light corresponding to multiple colors in the first color sequence may be a terminal device or a server.
  • the target device may be a terminal device, or a lamp connected to the terminal device that can emit various colors.
  • the terminal device can directly control the target device to sequentially emit light corresponding to multiple colors in the first color sequence; or, for example, the terminal device can By establishing a Bluetooth connection relationship with the lamp, or the terminal device and the lamp are located in the same wireless local area network, a communication connection relationship is established, thereby controlling the lamp to sequentially emit light corresponding to multiple colors in the first color sequence.
  • the server can send a control instruction to the target device based on the first color sequence.
  • the control instruction is used to control the target device to sequentially send out multiple colors corresponding to the first color sequence.
  • the target device sequentially emits light corresponding to multiple colors in the first color sequence.
  • the server sends a second control instruction to the terminal device based on the first color sequence.
  • the terminal device After receiving the control instruction, the terminal device sends out multiple colors corresponding to the first color sequence. corresponding light.
  • a Bluetooth connection may be established between the terminal device and the lamp capable of emitting various colors, or a Bluetooth connection may be established between the server and the lamp.
  • the communication relationship is connected; the server sends a control instruction to the terminal based on the first color sequence, where the control instruction carries the color identification information of the light emitted by the control lamp.
  • the terminal device After the terminal device receives the control instruction, it controls the lamp to emit light of the corresponding color.
  • the server can directly send a control instruction to the lamp; after receiving the control instruction, the lamp emits light of the color indicated by the color identification information in the control instruction.
  • each type of light can continue to emit light for a certain duration; within the emitting duration, the image acquisition device connected to the terminal device acquires the face image of the target face under the light.
  • the face image acquired under each light can be one. For example, based on the first color sequence including: red, green and blue; each color lasts for 2 seconds, and an image is acquired within 2 seconds; the specific process is: control the target device to emit red light for 2 seconds , and within 2 seconds of emitting red light, obtain the first face image corresponding to the red light; then control the target device to emit green light, and within 2 seconds of the target device emitting green light, obtain 1 image corresponding to the green light. The corresponding first face image; then control the target device to emit blue light, and the target device emits blue light for 2 seconds, and within 2 seconds of emitting blue light, obtain the first image corresponding to the blue light. Face images.
  • the specific process is: control the target device to emit red light for 5 seconds, and within 5 seconds of emitting the red light, acquire one first face image corresponding to the red light every second, and acquire a total of 5 images under the red light; Then control the target device to emit green light. Within 5 seconds of the target device emitting green light, obtain one first face image corresponding to the green light every second, and obtain a total of 5 images under green light; then control the target device to emit green light.
  • Blue light when the target device emits blue light for 5 seconds, and within 5 seconds of emitting blue light, acquire 1 first face image corresponding to blue light every second, and obtain a total of 1 face image under blue light 5 images.
  • the method when controlling the target device to emit light corresponding to multiple colors in the first color sequence, includes: detecting whether the target face is located in the light coverage area after the target device emits light; in response to the target face being located in the light coverage area, controlling the target device to sequentially emit the first color sequence.
  • the various colors in the light correspond to the light.
  • the infrared sensor can be used to detect whether the target face is located in the area covered by the target device emitting light.
  • the target device when it is detected that the target face is located in the light coverage area, the target device will be controlled to sequentially emit light corresponding to multiple colors in the first color sequence, so that the first face image corresponding to the color can be obtained. ; When it is detected that the target face is not located within the light coverage area, the target device cannot sequentially emit light corresponding to multiple colors in the first color sequence, and cannot obtain the first face image corresponding to the color. , in this case, when it is detected that the target face is not located in the light coverage area, a voice prompt can be provided that the user is within the light coverage area.
  • the image can also be obtained through an image acquisition device, and a target detection algorithm can be used to detect whether there is a face in the image; the target detection algorithm includes, for example, a face detection algorithm based on a neural network. After detecting the presence of a human face in the acquired image, it is determined that the target face exists in the area covered by the stored light.
  • color detection, liveness detection and identity detection can be performed on the first face images corresponding to each color.
  • an embodiment of the present disclosure provides a specific example of performing color detection on the first face image based on the first color sequence, including:
  • the first face image is obtained when photographing the face, and the color of each pixel in the first face image is sampled at the same time.
  • the color of each pixel has four channels (R, G, B, Alpha), based on (R, G, B, Alpha) of each pixel of the first face image, obtain the color information of the first face image corresponding to each color in the first color sequence; The color information can be used to determine whether it corresponds to each color in the first color sequence.
  • the light color emitted by the target device causes each first face image to be marked with a color corresponding to the light color.
  • the color marked on each first face image is the same as If the color of the light emitted by the target device corresponds, then the first face image matches each color in the first color sequence successfully.
  • the time when the target device emits each light color is preset, and the first face image will be acquired during the time when the target device emits each light color. For example, the time when the target device emits each light color lasts for 5 seconds. Then 5 face images can be obtained within these five seconds, that is, the timestamp of each first face image can be 1 second; compare the timestamp of the first face image with the time when the target device emits light.
  • Matching determining whether the colors of multiple first face images obtained during the time when the target device emits each light color match the color of the first face image when the color is detected; if the target device emits each light color If the colors of the multiple first face images acquired within a period of time are the same as the colors used when performing color detection on the first face image, then the matching is successful.
  • the color information of the first face image corresponding to each color is matched. If the matching is successful, the first face image passes the color detection.
  • an embodiment of the present disclosure provides a specific example of performing liveness detection on a first face image, including:
  • S301 For each color in the first color sequence, determine the first target face image corresponding to each color from the first face images corresponding to each color.
  • determining the first target face image corresponding to each color includes: determining the quality information of the first face image corresponding to each color; determining based on the quality information, The first target face image corresponding to each color.
  • the quality information of the first face image includes, for example, at least one of the following: the clarity of the first face image, the size of the first face in the first face image, the first face in the first face image. of completeness.
  • the quality score of the first face image is higher.
  • This score represents the quality of the first face image, and can be determined as needed.
  • the sharpness size changes; if the quality score of the first face image is high, the first face image corresponding to each color is the first target face image; for example, the system's first face image sharpness preset Suppose the threshold is 100.
  • the clarity of the first face image corresponding to red in the first color sequence is 100, then the quality score is 100, and the first face image corresponding to red is the first target face image. ; If the clarity of the first face image corresponding to red in the first color sequence is 60, then the quality score is 60, then the first face image corresponding to red is not the first target face image.
  • the size of the first face in the first face image can be detected by passing the first face image through a fully convolutional network to obtain a feature map. Each "point" on the feature map corresponds to the position mapped to the first person.
  • the area belonging to the first face in the face image, the first face area in the first face image is regarded as the face candidate frame, and the first face in the first face image is determined based on the size of the face candidate frame size.
  • the size of the user's face is constant, when the first face image is obtained, the size of the user's face will be calculated to obtain the user's preset face size threshold. For example, when the size of the first face in the first face image corresponding to each color reaches the preset face size threshold, the quality score of the first face image is higher, and the score represents the first face size.
  • the quality of the face image if the quality score of the first face image is high, the first face image corresponding to each color is the first target face image; for example, in the first face image of the system
  • the preset threshold for the size of the first face is 150*150.
  • the quality score can be 100, then the first face image corresponding to red is the first target face image; if the size of the first face in the first face image corresponding to red in the first color sequence is 100*100, Then the quality score is 50, and the first face image corresponding to the red color is not the first target face image.
  • the completeness of the first face in the first face image can be detected by the number of face key points. Based on the facial key point detection model, the facial key points are detected and the detection results of the facial key points are obtained. For example, there are 13 facial key points that can be detected, and when detecting a certain face image, 5 key points are detected, which can indicate that the face is incomplete.
  • the integrity of the first face in the first face image is judged based on the number of face key points; the number of face key points detected by the face key point detection model, and the actual number of face key points detected by the first face image The larger the difference between the quantities, the lower the completeness.
  • the preset number of key points of the front face in the face image is 100.
  • face key points are detected on the front face of the first face in the first face image.
  • the integrity of the first face in the first face image is 100 points; when the number of face key points is between 80 and 90, the first person in the first face image
  • the face completeness is 80 points; when the number of face key points is between 70 and 80, the first face completeness in the first face image is 70 points; when the number of face key points is between 60 and 70 time, the completeness of the first face in the first face image is 60 points; the fewer key points of the first face in the first face image, the lower the score.
  • the quality score of the first face image is higher, and the score represents the quality of the first face image.
  • the quality score of the first face image is high, then the first face image corresponding to each color is the first target face image; for example, the completeness of the first face in the first face image is 100 points, then the quality score can be 100, then the first face image corresponding to the red color is the first target face image; if the completeness of the first face image corresponding to the red color in the first color sequence is 60 time sharing, the quality score is 60, and the first face image corresponding to the red color is not the first target face image.
  • S302 Perform life detection on the first target face image corresponding to each color, and obtain the life detection result of the first target face image corresponding to each color.
  • each color in the first color sequence has a one-to-one corresponding first face image
  • the first target face image is subjected to in vivo imaging. Detection, based on the liveness detection result of the corresponding first target face image, determines whether the first target face image is a live face.
  • the optical flow field method can be used for live detection. This method mainly uses the time domain changes and correlations of the pixel intensity data in the image sequence to determine the "movement" of the respective pixel positions. From the image sequence The operating information of each pixel is obtained, and Gaussian difference filter, local binary pattern (Local Binary Patterns, LBP) texture features and support vector machine are used for statistical analysis of data.
  • LBP Local Binary Patterns
  • the optical flow field is sensitive to object movement, and the optical flow field can be used to uniformly detect eye movement and blinking; since the real face is not absolutely static, there are micro-expressions, such as the rhythm of the eyelids and eyeballs, blinking, lips and surrounding cheeks. Scaling, etc., and judging whether it is a living face based on the operating information of each pixel.
  • determining that the first face image passes the live body detection includes: responding to a live body detection result indicating that the target face is a live face and the number of first target face images reaches a predetermined number. Assuming that the quantity and the quality information corresponding to the first target face image that the live body detection result indicates that the target face is a live face meets the preset conditions, it is determined that the first face image passes the live body detection.
  • the liveness detection result indicates that the number of first target face images whose target face is a live face reaches a preset number.
  • the preset number of live face images is 3.
  • the target face is the first target face of a living face
  • the first face image meets the condition of reaching the preset number.
  • the quality information corresponding to the first target face image in which the live body detection result indicates that the target face is a live face satisfies the preset conditions.
  • the preset conditions may include, for example, any of the following b1 to b3:
  • the first target face image quality score is the highest when the target face is a living face.
  • the target face is the first target face image of a living human face.
  • the corresponding quality information meets the preset conditions.
  • the target face with a quality score of 89 is the first target face of a living face.
  • the quality information corresponding to the image meets the preset conditions; the quality information corresponding to the first target face image in which the other target faces are living faces does not meet the preset conditions.
  • the quality score of the first target face image whose target face is a living face is greater than the preset quality score threshold.
  • the system's preset quality score threshold for the first face image is 80 points. If the target face is a living face and the quality score of the first target face image is 85 points, then the target face is a living face. The quality information corresponding to the first target face image meets the preset conditions; if the quality score of the first target face image of the target face is a living face, then the quality score of the first target face image of the target face is a living face, then the quality score of the first target face image of the target face is a living face. The quality information corresponding to a target face image does not meet the preset conditions.
  • the number of quality scores of the first target face images whose target faces are living faces is greater than the preset quality score threshold is greater than the preset number.
  • the system's preset quality score threshold for the first face image is 80 points, and the preset number of living faces is 3; if the target face is a living face, the quality score of the first target face image is 89 points. There are 5 images in total, then the quality information corresponding to the first target face image of a living human face satisfies the preset conditions.
  • the first target face image passes the live body detection.
  • an embodiment of the present disclosure provides a specific example of identity detection based on a first face image.
  • the steps include S401 to S402, where:
  • the first face image with better quality information is selected as the second target from the first face image that has passed the live body detection.
  • the determination method is similar to the above-mentioned S301, and will not be described again here.
  • S402 Perform identity detection based on the second target face image and the pre-stored face registration image.
  • the pre-stored face registration image is a face registration image stored in the database after receiving the face registration request.
  • feature extraction can be performed on the second target face image first to obtain the first feature information of the second target face image; and Feature extraction is performed on the pre-stored face registration image to obtain the second feature information of the face registration image, and then the distance between the first feature information and the second feature information is determined. According to the distance, it is determined whether the second target face image and the face registration image belong to the same user. If both belong to the same user, the identity detection is passed. If they belong to different users, the identity test fails.
  • the second target face image is matched with a pre-stored face registration image, and the matching score can be determined based on the distance. If the matching score is greater than the preset matching score threshold, the second target face image is determined to be the same as the prestored face registration image; for example, the preset matching score threshold is 85, and when the second target face image is the same as the prestored face registration image, If the matching score of the face registration image is 90, then the second target face image is determined to be the same as the pre-stored face registration image, that is, it is the same person; when the matching score of the second target face image and the pre-stored face registration image is 80, Then the second target face image is determined to be different from the pre-stored face registration image, that is, it is not the same person.
  • the preset matching score threshold is 85, and when the second target face image is the same as the prestored face registration image, If the matching score of the face registration image is 90, then the second target face image is determined to be the same as the pre-stored
  • embodiments of the present disclosure also provide a specific method for obtaining a face registration image, as shown in Figure 5, which includes:
  • S501 In response to receiving a face registration request, obtain a second color sequence; the second color sequence includes: multiple colors;
  • the acquisition method of the second color sequence is similar to the acquisition method of the first color sequence.
  • S101 the acquisition method of the first color sequence.
  • the method of obtaining the second face image is similar to the method of obtaining the first face image.
  • the method of obtaining the second face image is similar to the method of obtaining the first face image.
  • S503 Based on the second color sequence, perform color detection on the second face image, and perform life detection on the second face image.
  • the method of performing color detection and liveness detection on the second face image based on the second color sequence is similar to the above-mentioned method of performing color detection and liveness monitoring on the first face image based on the first color sequence.
  • the embodiment of S103 will not be described again here.
  • S504 In response to the second face image passing the color detection and the living body detection, determine the face registration image based on the second face image.
  • the face registration image can be determined from the second face image according to the image quality of the second face image, and the face registration image can be stored in the database using for identity testing.
  • Target interaction requests include: reservation requests, and perform interactive operations corresponding to the reservation requests, including:
  • course reservation information Obtain course reservation information, make course reservations based on the course reservation information, and generate course reservation records.
  • the course reservation operation can be performed.
  • a course reservation record will be generated; the course reservation record indicates that the user can enter the driving school to study within the course reservation time.
  • Target interaction requests include: class registration requests, and perform interactive operations corresponding to class registration requests, including:
  • the security inspection equipment In response to the course reservation record indicating that the reserved course start time meets the preset time conditions, the security inspection equipment is controlled to open the release channel.
  • obtaining the course reservation record means obtaining the user's reservation time and reserved courses, indicating whether the reserved course start time meets the preset time conditions through the course reservation record, and controlling whether the security inspection equipment opens the release channel; where the course reservation record includes the course The start time and the current time the user made the reservation.
  • the preset time conditions may include any of the following conditions (3) to (5), among which:
  • the reserved course start time is one day from the current time. If the reserved course start time is one day from the current time, then the reserved course start time meets the preset time conditions and controls the opening of the security inspection equipment. Through the release channel, users can enter the driving school to study; for example, if the reserved course start time is 2:00 pm on September 1, and the user reserves a course on September 1, the user can enter the driving school to study.
  • the time difference between the reserved course start time and the current time is determined. If the time difference between the reserved course start time and the current time is less than the preset time difference threshold, the reserved course If the start time meets the preset time conditions, the security inspection equipment is controlled to open the release channel, and the user can enter the driving school to study; for example, the scheduled start time of the course is 14:00 on September 1st, and the user is required to start at 8:00 to 14:00 on September 1st.
  • Course reservations can be made within 6 hours of the course start time. Users can make course reservations at 9:00 on September 1 and enter the driving school to study.
  • the reserved course start time is located after the current time, and the time difference between the reserved course start time and the current time is less than the preset time difference threshold.
  • the reserved course start time is located after the current time, and the time difference between the reserved course start time and the current time is determined. If the reserved course start time and the current time are If the time difference is less than the preset time difference threshold, the scheduled start time of the course meets the preset time conditions, the security inspection equipment is controlled to open the release channel, and the user can enter the driving school to study; for example, the scheduled start time of the course is September 1st, and the user needs to be in the preset time difference.
  • Course reservations must be made within the time difference threshold, that is, within 7 days from the course start time, the reservation can be successfully made. If the user makes a reservation for the course on August 25, the course can be successfully reserved, and the user can enter the driving school to study on September 1.
  • the reserved course start time is before the current time, and the time difference between the reserved course start time and the current time is less than the preset time difference threshold.
  • the reserved course start time is before the current time, and the time difference between the reserved course start time and the current time is determined. If the reserved course start time and the current time are If the time difference is less than the preset time difference threshold, then the reserved course start time meets the preset time conditions, the security inspection equipment is controlled to open the release channel, and the user can enter the driving school to study; for example, the reserved course start time is September 1, and the user can enter the driving school after the course starts. If you reserve a course within 3 days, if you reserve a course on September 2, you can successfully reserve a course and enter the driving school to study.
  • the target interaction request includes: vehicle practical operation request, and performs the interactive operation corresponding to the vehicle practical operation request, including:
  • the user information In response to the user information entering the queuing state reaching the queuing end condition, the user information is assigned a target vehicle for vehicle operation practice.
  • the user information corresponding to the course registration record can be controlled to enter the queuing state.
  • the user information entering the queuing state reaches the queuing end condition, the user can perform vehicle operation practice.
  • the queuing end condition can be that the user's reserved vehicle operation practice time arrives, and the user can perform vehicle operation practice; for example, the vehicle operation time reserved by user A is 3 o'clock in the afternoon, and when the time of the day reaches 3 o'clock in the afternoon, user A can Learn how to operate a vehicle.
  • the condition for the end of the queue can also be that the vehicle operation practice of other users in front of the user ends, and the user can perform vehicle operation practice; for example, after the vehicle operation practice of user B in front of user A ends, user A can perform vehicle operation practice.
  • the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process.
  • the specific execution order of each step should be based on its function and possible The internal logic is determined.
  • the embodiment of the present disclosure also provides an interactive control device corresponding to the interactive control method. Since the principle of solving the problem of the device in the embodiment of the present disclosure is similar to the above-mentioned interactive control method in the embodiment of the present disclosure, the implementation of the device Please refer to the implementation of the method, and the repeated parts will not be repeated.
  • the device includes: an acquisition module 61, a detection module 62, and a processing module 63; wherein,
  • Obtaining module 61 is configured to obtain a first color sequence in response to obtaining the target interaction request; the first color sequence includes: multiple colors; based on the first color sequence, obtain the target face in the first color sequence The corresponding first face images in multiple colors;
  • the detection module 62 is configured to perform color detection and life detection on the first face image based on the first color sequence, and perform identity detection based on the first face image;
  • the processing module 63 is configured to perform an interactive operation corresponding to the target interaction request in response to the first face image passing the color detection, the living body detection, and the identity detection.
  • the acquisition module 61 acquires the first face images corresponding to the target face in multiple colors in the first color sequence based on the first color sequence. , used to control the target device to sequentially emit light corresponding to multiple colors in the first color sequence based on the first color sequence, and when controlling the target device to emit light of each color, obtain The first face image corresponding to this color.
  • the detection module 62 is used to detect whether the target face is the same when the control target device emits light corresponding to multiple colors in the first color sequence in sequence. Located within the light coverage area after the target device emits light; in response to the target face being located within the light coverage area, controlling the target device to sequentially emit light corresponding to multiple colors in the first color sequence .
  • the detection module 62 when performing color detection on the first face image based on the first color sequence, is configured to detect each color in the first color sequence. Color, perform color detection on the first face image corresponding to the color, and obtain the color information of the first face image corresponding to the color; match each color with the color information of the corresponding first face image; In response to the multiple colors in the first color sequence successfully matching the color information corresponding to the first face image, it is determined that the first face image passes the color detection.
  • the detection module 62 is configured to detect, for each color in the first color sequence, from each color of the first face image when performing life detection on the first face image. Among the first face images corresponding to each color, determine a first target face image corresponding to each color; perform in vivo detection on the first target face image corresponding to each color to obtain each color The corresponding live body detection result of the first target face image; in response to the live body detection result indicating that the number of first target face images of the target face being a live face reaches a preset number, determining the first face image Through the in vivo detection.
  • the method further includes a determination module 64, used to determine the first target face image corresponding to each color from the first face image corresponding to each color. Determine the quality information of the first face image corresponding to each color; determine the first target face image corresponding to each color based on the quality information.
  • a determination module 64 used to determine the first target face image corresponding to each color from the first face image corresponding to each color. Determine the quality information of the first face image corresponding to each color; determine the first target face image corresponding to each color based on the quality information.
  • the determination module 64 determines that the number of first target face images indicating that the target face is a live face in response to the live body detection result reaches a preset number.
  • the number of first target face images used to respond to the live body detection result indicating that the target face is a live human face reaches a preset number, and the live body detection result indicates that the If the quality information corresponding to the first target face image in which the target face is a living face meets the preset conditions, it is determined that the first target face image passes the living body detection.
  • the detection module 62 when performing identity detection based on the first face image, is configured to detect the first face corresponding to each color in the first color sequence based on The second target face image is determined based on the image quality information; and identity detection is performed based on the second target face image and the pre-stored face registration image.
  • the detection module 62 is configured to obtain a second color sequence in response to receiving a face registration request; the second color sequence includes: multiple colors; based on the second color sequence, Obtain second face images corresponding to the target face in multiple colors in the second color sequence; perform color detection on the second face image based on the second color sequence, and perform color detection on the second face image.
  • the second face image is subjected to life detection; in response to the second face image passing the color detection and the life detection, the face registration image is determined based on the second face image.
  • the target interaction request includes at least one of the following:
  • the target interaction request includes: a reservation request
  • the processing module 63 is used to obtain course reservation information when performing an interaction operation corresponding to the target interaction request, and based on Use the course reservation information to make course reservations and generate course reservation records.
  • the target interaction request includes: a class registration request
  • the processing module 63 is used to obtain course reservation records when performing an interaction operation corresponding to the target interaction request; respond
  • the course reservation record indicates that the scheduled start time of the course meets the preset time conditions, the security inspection equipment is controlled to open the release channel.
  • the target interaction request includes: a vehicle practice operation request, and the processing module 63 is used to obtain course registration records when performing the interaction operation corresponding to the target interaction request; Based on the course registration record, the user information corresponding to the course registration record is controlled to enter the queuing state; in response to the user information entering the queuing state reaching the queuing end condition, the user information is assigned a target vehicle for vehicle operation practice.
  • An embodiment of the present disclosure also provides a computer device. As shown in Figure 7, which is a schematic structural diagram of the computer device provided by an embodiment of the present disclosure, it includes:
  • Processor 71 and memory 72 stores machine-readable instructions executable by the processor 71, and the processor 71 is used to execute the machine-readable instructions stored in the memory 72, and the machine-readable instructions are used by the processor 71 When executing, processor 71 performs the following steps:
  • the first color sequence includes: multiple colors
  • an interaction operation corresponding to the target interaction request is performed.
  • the above-mentioned memory 72 includes a memory 721 and an external memory 722; the memory 721 here is also called internal memory, and is used to temporarily store the operation data in the processor 71, as well as the data exchanged with external memory 722 such as a hard disk.
  • the processor 71 communicates with the computer through the memory 721.
  • External memory 722 performs data exchange.
  • Embodiments of the present disclosure also provide a computer-readable storage medium.
  • the computer-readable storage medium stores a computer program. When the computer program is run by a processor, the steps of the interactive control method described in the above method embodiment are executed.
  • the storage medium may be a volatile or non-volatile computer-readable storage medium.
  • Embodiments of the present disclosure also provide a computer program product.
  • the computer program product carries program code.
  • the instructions included in the program code can be used to execute the steps of the interactive control method described in the above method embodiment. For details, please refer to the above method. The embodiments will not be described again here.
  • the above-mentioned computer program product can be specifically implemented by hardware, software or a combination thereof.
  • the computer program product is embodied as a computer storage medium.
  • the computer program product is embodied as a software product, such as a Software Development Kit (SDK), etc. wait.
  • SDK Software Development Kit
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in various embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor.
  • the technical solution of the present disclosure is essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Image Processing (AREA)

Abstract

本公开提供了一种交互控制方法、装置、计算机设备及存储介质，其中，该方法包括：响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。

Description

交互控制方法、装置、计算机设备及存储介质
本公开要求于2022年03月24日提交中国专利局、申请号为202210303610.3、发明名称为“交互控制方法、装置、计算机设备及存储介质”，其全部内容通过引用结合在本公开中
技术领域
本公开涉及身份识别技术领域，具体涉及一种交互控制方法、装置、计算机设备及存储介质。
背景技术
随着道路交通的不断发展，我国拥有车辆的居民也越来越多。道路交通安全的保障，在很大程度上依赖于居民驾驶车辆的能力。驾校作为居民学习驾驶车辆的场所，在居民在获得驾驶证前，需要保证居民进行充分的相关法规学习、和车辆驾驶练习。而在驾校的购买课程、预约课程、排队登记等场景中，人流量较大且当前主要通过人工识别身份证来核实用户信息，这样方式不仅耗时长，而且人眼的识别人脸存在准确度低，可能存在代课的行为，这种行为会造成一定的安全隐患。
发明内容
本公开实施例至少提供一种交互控制方法、装置、计算机设备及存储介质。
第一方面，本公开实施例提供了一种交互控制方法，该方法包括：响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。
这样，通过在某第一颜色序列下获取第一人脸图像，并基于获取的第一人脸图像进行颜色检测、活体检测以及身份检测，保证了发起目标交互请求的用户的身份，具有更高的识别精度、以及识别速度，提升在驾校中核实用户信息的效率以及识别用户的准确度，防止出现代课行为，降低由于代课行为所导致的安全隐患。
一种可选的实施方式中，所述基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像，包括：基于所述第一颜色序列，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，并在控制所述目标设备发射出每种颜色的光线时，获取与该种颜色对应的第一人脸图像。
这样，可以在与第一颜色序列对应的光线中获取第一人脸图像，降低第一人脸图像获取过程被假冒的可能，提升人脸检测的安全性。
一种可选的实施方式中，所述控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，包括：检测所述目标人脸是否位于所述目标设备发出光线后的光线覆盖区域内；响应于所述目标人脸位于所述光线覆盖区域内，控制目标设备依次发出与所述第一颜 色序列中的多种颜色分别对应的光线。
这样，通过检测目标人脸是否在目标设备发出光线后的光线覆盖区域内，确保获取的图像中包括了目标人脸，提升检测的效率。
一种可选的实施方式中，基于所述第一颜色序列，对所述第一人脸图像进行颜色检测，包括：针对所述第一颜色序列中的每种颜色，对于该颜色对应的第一人脸图像进行颜色检测，得到与该颜色对应的第一人脸图像的颜色信息；将所述每种颜色和对应第一人脸图像的颜色信息进行匹配；响应于所述第一颜色序列中的多种颜色分别与对应第一人脸图像的颜色信息匹配成功，确定所述第一人脸图像通过所述颜色检测。
这样，对第一颜色序列下每种颜色对应的第一人脸图像分别进行颜色的匹配，在每种颜色和对应的第一人脸图像的颜色信息成功的前提下，才确定第一人脸图像通过颜色检测，从而减少在人脸检测过程中的为即时获取的人脸，而并非是仿冒的人脸，保证获得的第一人脸图像的可靠性。
一种可选的实施方式中，所述对所述第一人脸图像进行活体检测，包括：针对所述第一颜色序列中的每种颜色，从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像；对所述每种颜色对应的第一目标人脸图像进行活体检测，得到所述每种颜色对应的第一目标人脸图像的活体检测结果；响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测。
这样，在对第一人脸图像进行活体检测时，活体检测结果指示目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，才会确定第一人脸图像通过活体检测，从而保证活体检测结果的准确性。
一种可选的实施方式中，所述从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像，包括：确定所述每种颜色对应的第一人脸图像的质量信息；基于所述质量信息，确定与所述每种颜色对应的第一目标人脸图像。
这样，通过每种颜色对应的第一人脸图像的质量信息，确定与每种颜色对应的第一目标人脸图像，使得确定的第一目标人脸图像在进行活体检测时，具有更高的检测精度。
一种可选的实施方式中，所述响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测，包括：响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量、且活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件，确定所述第一人脸图像通过所述活体检测。
这样，通过确定目标人脸为活体人脸的第一目标人脸图像的数量是否达到预设数量、且活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息是否满足预设条件，可以更为准确地确定第一人脸图像是否能够通过活体检测。
一种可选的实施方式中，所述基于所述第一人脸图像进行身份检测，包括：基于所述第一颜色序列中每种颜色对应的第一人脸图像的质量信息，确定第二目标人脸图像；基于所述第二目标人脸图像、以及预存的人脸登记图像进行身份检测。
这样，通过将第二目标人脸图像与预存的人脸登记图像进行检测，确保第二目标人脸图像与预存的第一人脸图像是否为同一人，提升人脸识别准确度，增加可靠性。
一种可选的实施方式中，还包括：响应于接收到人脸登记请求，获取第二颜色序列；所述第二颜色序列包括：多种颜色；基于第二颜色序列，获取目标人脸在所述第二颜色序列中的多种颜色下分别对应的第二人脸图像；基于所述第二颜色序列，对所述第二人脸图像进行 颜色检测，以及，对所述第二人脸图像进行活体检测；响应于所述第二人脸图像通过所述颜色检测、以及所述活体检测，基于所述第二人脸图像确定所述人脸登记图像。
这样，通过响应于接收到人脸登记请求，在某第一颜色序列下获取第二人脸图像，并基于获取的第二人脸图像进行颜色检测和活体检测，从通过颜色检测和活体检测的第二人脸图像来确定人脸登记图像，使得对第一人脸图像进行身份检测时，提供了人脸登记图像，从而在身份检测过程中具有更高的识别精度、以及识别速度，增加可靠性。
一种可选的实施方式中，所述目标交互请求包括下述至少一种：登录请求、课程购买请求、预约请求、上课登记请求、车辆实践操作请求。
一种可选的实施方式中，所述目标交互请求包括：预约请求，所述执行与所述目标交互请求对应的交互操作，包括：
获取课程预约信息，并基于所述课程预约信息进行课程预约，生成课程预约记录。
这样，通获取课程预约信息，可以准确判断用户预约了哪些课程以及课程的开始时间，便于后续用户进入驾校学习时控制安检设备开启放行通道。
一种可选的实施方式中，所述目标交互请求包括：上课登记请求，所述执行与所述目标交互请求对应的交互操作，包括：获取课程预约记录；响应于所述课程预约记录指示预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道。
这样，通过获取到课程的预约记录，通过课程预约记录确定进入驾校学习的用户是否与预约课程用户是否为同一人，防止出现代课行为。
一种可选的实施方式中，所述目标交互请求包括：车辆实践操作请求，所述执行与所述目标交互请求对应的交互操作，包括：获取课程登记记录；基于所述课程登记记录，控制与所述课程登记记录对应的用户信息进入排队状态；响应于进入排队状态的用户信息达到排队结束条件，为所述用户信息分配用于车辆操作实践的目标车辆。
这样，通过获取课程登记记录，控制与课程登记记录对应的用户信息进入排队状态，当进入排队状态的用户信息达到排队结束条件时，用户才可以进行车辆操作实践，确保进行车辆操作实践的用户与课程登记的用户为同一人，防止出现代课行为。
第二方面，本公开实施例还提供一种交互控制装置，包括：
获取模块，用于响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；
检测模块，用于基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；
处理模块，用于响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。
一种可选的实施方式中，所述获取模块，在所述基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像时，用于基于所述第一颜色序列，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，并在控制所述目标设备发射出每种颜色的光线时，获取与该种颜色对应的第一人脸图像。
一种可选的实施方式中，所述检测模块，在所述控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线时，用于检测所述目标人脸是否位于所述目标设备发出光线后的光线覆盖区域内；响应于所述目标人脸位于所述光线覆盖区域内，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线。
一种可选的实施方式中，所述检测模块，在基于所述第一颜色序列，对所述第一人脸图像进行颜色检测时，用于针对所述第一颜色序列中的每种颜色，对于该颜色对应的第一人脸图像进行颜色检测，得到与该颜色对应的第一人脸图像的颜色信息；将所述每种颜色和对应第一人脸图像的颜色信息进行匹配；响应于所述第一颜色序列中的多种颜色分别与对应第一人脸图像的颜色信息匹配成功，确定所述第一人脸图像通过所述颜色检测。
一种可选的实施方式中，所述检测模块，在所述对所述第一人脸图像进行活体检测时，用于针对所述第一颜色序列中的每种颜色，从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像；对所述每种颜色对应的第一目标人脸图像进行活体检测，得到所述每种颜色对应的第一目标人脸图像的活体检测结果；响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测。
一种可选的实施方式中，还包括确定模块，在所述从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像时，用于确定所述每种颜色对应的第一人脸图像的质量信息；基于所述质量信息，确定与所述每种颜色对应的第一目标人脸图像。
一种可选的实施方式中，所述确定模块，在所述响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测时，用于响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量、且活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件，确定所述第一人脸图像通过所述活体检测。
一种可选的实施方式中，所述检测模块，在所述基于所述第一人脸图像进行身份检测时，用于基于所述第一颜色序列中每种颜色对应的第一人脸图像的质量信息，确定第二目标人脸图像；基于所述第二目标人脸图像、以及预存的人脸登记图像进行身份检测。
一种可选的实施方式中，所述检测模块，用于响应于接收到人脸登记请求，获取第二颜色序列；所述第二颜色序列包括：多种颜色；基于第二颜色序列，获取目标人脸在所述第二颜色序列中的多种颜色下分别对应的第二人脸图像；基于所述第二颜色序列，对所述第二人脸图像进行颜色检测，以及，对所述第二人脸图像进行活体检测；响应于所述第二人脸图像通过所述颜色检测、以及所述活体检测，基于所述第二人脸图像确定所述人脸登记图像。
一种可选的实施方式中，所述目标交互请求包括下述至少一种：
登录请求、课程购买请求、预约请求、上课登记请求、车辆实践操作请求。
一种可选的实施方式中，所述目标交互请求包括：预约请求，所述处理模块，在所述执行与所述目标交互请求对应的交互操作时，用于获取课程预约信息，并基于所述课程预约信息进行课程预约，生成课程预约记录。
一种可选的实施方式中，所述目标交互请求包括：上课登记请求，所述处理模块，在所述执行与所述目标交互请求对应的交互操作时，用于获取课程预约记录；响应于所述课程预约记录指示预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道。
一种可选的实施方式中，所述目标交互请求包括：车辆实践操作请求，所述处理模块，在所述执行与所述目标交互请求对应的交互操作时，用于获取课程登记记录；基于所述课程登记记录，控制与所述课程登记记录对应的用户信息进入排队状态；响应于进入排队状态的用户信息达到排队结束条件，为所述用户信息分配用于车辆操作实践的目标车辆。
第三方面，本公开可选实现方式还提供一种计算机设备，处理器、存储器，所述存储器存储有所述处理器可执行的机器可读指令，所述处理器用于执行所述存储器中存储的机器可 读指令，所述机器可读指令被所述处理器执行时，所述机器可读指令被所述处理器执行时执行上述第一方面，或第一方面中任一种可能的实施方式中的步骤。
第四方面，本公开可选实现方式还提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被运行时执行上述第一方面，或第一方面中任一种可能的实施方式中的步骤。
关于上述交互控制装置、计算机设备、及计算机可读存储介质的效果描述参见上述交互控制方法的说明，这里不再赘述。
为使本公开的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。
附图说明
为了更清楚地说明本公开实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，此处的附图被并入说明书中并构成本说明书中的一部分，这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开的技术方案。应当理解，以下附图仅示出了本公开的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。
图1示出了本公开实施例所提供的一种交互控制方法的流程图；
图2示出了本公开实施例所提供的一种对所述第一人脸图像进行颜色检测的具体示例的流程图；
图3示出了本公开实施例所提供的一种对第一人脸图像进行活体检测的具体示例的流程图；
图4示出了本公开实施例所提供的一种基于第一人脸图像进行身份检测的具体示例的流程图；
图5示出了本公开实施例所提供的一种获取人脸登记图像的具体方法的流程图；
图6示出了本公开实施例所提供的一种交互控制装置的示意图；
图7示出了本公开实施例所提供的一种计算机设备的示意图。
具体实施方式
为使本公开实施例的目的、技术方案和优点更加清楚，下面将结合本公开实施例中附图，对本公开实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本公开一部分实施例，而不是全部的实施例。通常在此处描述和示出的本公开实施例的组件可以以各种不同的配置来布置和设计。因此，以下对本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围，而是仅仅表示本公开的选定实施例。基于本公开的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本公开保护的范围。
基于背景技术中的相关问题，本公开提供了一种交互控制方法，通过在某第一颜色序列下获取第一人脸图像，并基于获取的第一人脸图像进行颜色检测、活体检测以及身份检测，保证了发起目标交互请求的用户的身份，具有更高的识别精度、以及识别速度，提升在驾校中核实用户信息的效率以及识别用户的准确度，防止出现代课行为，降低由于代课行为所导致的安全隐患。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附 图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。
为便于对本实施例进行理解，首先对本公开实施例所公开的一种交互控制方法进行详细介绍，本公开实施例所提供的交互控制方法的执行主体一般为具有一定计算能力的计算机设备，该计算机设备例如包括：终端设备或服务器或其它处理设备，终端设备可以为计算机设备，计算机设备例如电脑，平板电脑等。在一些可能的实现方式中，该交互控制方法可以通过处理器调用存储器中存储的计算机可读指令的方式来实现。
本公开实施例提供的交互控制方法，除了能够应用于驾校场景下，还可以用于其他场景，如银行系统、门禁系统等，本公开实施例不对应用的具体场景进行限定。
下面对本公开实施例提供的交互控制方法加以说明。
参见图1所示，为本公开实施例提供的交互控制方法的流程图，所述方法包括步骤S101～S104，其中：
S101：响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；
S102：基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；
S103：基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；
S104：响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。
本公开实施例主要通过响应于获取到目标交互请求，获取第一颜色序列，基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测，当所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测后，才执行与目标交互请求对应的交互操作，保证了发起目标交互请求的用户的身份，具有更高的识别精度、以及识别速度，提升在驾校中核实用户信息的效率以及识别用户的准确度，防止出现代课行为，降低由于代课行为所导致的安全隐患。
针对上述S101：
目标交互请求例如包括下述至少一种：登录请求、课程购买请求、预约请求、上课登记请求、车辆实践操作请求。
其中，以将本公开实施例提供的交互控制方法应用于驾校场景为例，在终端设备上，可以部署与驾校对应的应用程序；用户通过该应用程序，可以进入登录页面，并通过登录页面发起登录请求。其中，终端设备可以是用户手持的移动终端；也可以是在某地点的专用终端设备。例如在驾校的服务大厅设置的专用驾校课程服务设备，通过该驾校课程服务设备，用户可以自助进行注册、登录、购买课程等操作。
课程购买请求，是通过应用程序的服务页面发起的购买驾校课程的请求。
本公开实施例提供的交互控制方法可以应用于终端设备，也可以应用于服务器。终端设备或者服务器可以响应于获取到目标交互请求，获取第一颜色序列，
其中，在交互控制方法应用于终端设备中的情况下，用户通过终端设备触发目标交互请求；终端设备响应于用户触发的目标交互请求，向服务器发送获取第一颜色序列的请求，服务器接收到终端设备发送的获取第一颜色序列的请求，基于预设的颜色序列的生成规则，生成第一颜色序列，并向终端设备返回该第一颜色序列。
在交互控制方法应用于服务器的情况下，服务器能够接收终端设备发起的目标交互请求。服务器在接收到终端设备发起的目标交互请求后，直接按照预设的颜色序列的生成规则，生成第一颜色序列。
其中，颜色序列的生成规则，例如包括：颜色序列中可选的颜色的类型限制条件、颜色序列中的颜色的数量限制条件。
例如，可选的颜色包括：红色、绿色、黄色、蓝色、紫色、橙色、粉色；
颜色序列中的颜色的数量为5。
则可以在上述可选的颜色中随机选择5种，并将随机选择的5种颜色随机排列，构成第一颜色序列。
颜色序列的生成规则可以根据实际的需要进行设置，本公开实施例中不做限定。其中：
不同目标交互请求获取的颜色序列中的颜色可以相同(此时，不同颜色序列中各颜色的位置有所区别)。示例性的，针对不同目标交互请求对应第一颜色的颜色包括相同颜色、但颜色位置不同的情况，例如，第一次响应于获取到目标交互请求后，第一次获取的第一颜色序列为：红色、黄色、绿色；第二次响应于获取到目标交互请求后，第二次获取的第一颜色序列为：绿色、红色、黄色。
不同目标交互请求获取的颜色序列中的颜色是完全不相同的情况下，当响应于获取到目标交互请求后，获取的第一颜色序列中的颜色可以不相同，例如，第一次响应于获取到目标交互请求后，第一次获取的第一颜色序列为红色、黄色、绿色；第二次响应于获取到目标交互请求后，第二次获取的第一颜色序列为白色、紫色、橙色。
不同目标交互请求获取的颜色序列中的颜色不全相同的情况下，当响应于获取到目标交互请求后，获取的第一颜色序列中的颜色可以存在至少一种颜色不相同；例如，第一次响应于获取到目标交互请求后，第一次获取的第一颜色序列为红色、黄色、绿色；第二次响应于获取到目标交互请求后，第二次获取的第一颜色序列为红色、黄色、蓝色。
针对上述S102：本公开实施例提供了一种所述基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像的具体方式，包括：
基于所述第一颜色序列，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，并在控制所述目标设备发射出每种颜色的光线时，获取与该种颜色对应的第一人脸图像。
目标人脸是图像采集设备需要采集的用户的活体人脸，不可以是用户的照片、用户的人脸图像视频，防止通过照片或者视频假冒人脸信息，造成一定的安全隐患。
基于第一颜色序列，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，使得在多种颜色光线下的人脸反射的光线有一一对应的颜色，从而可以获取与多种颜色光线分别对应的第一人脸图像。
控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线的控制端可以是终端设备也可以是服务器。
其中，目标设备可以是终端设备，也可以是与终端设备连接的能够发射各种颜色的灯。
在交互控制方法应用于终端设备的情况下，基于第一颜色序列，终端设备可以直接控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线；或者，例如终端设备可以通过与灯建立蓝牙连接关系，或者终端设备和灯位于同一个无线局域网中，建立通信连接关系，从而控制灯依次发出与所述第一颜色序列中的多种颜色分别对应的光线。
在交互控制方法应用于服务器的情况下，服务器可以基于第一颜色序列，向目标设备发 送控制指令，所述控制指令用来控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，目标设备接收服务器发送的控制指令后，依次发出与所述第一颜色序列中的多种颜色分别对应的光线。
例如，当目标设备为终端设备的情况下，服务器基于第一颜色序列，向终端设备发送第二控制指令，终端设备接收到控制指令后，发出与所述第一颜色序列中的多种颜色分别对应的光线。
又例如，当目标设备为能够发射各种颜色的灯的情况下，其中，可以是终端设备和与能够发射各种颜色的灯之间可以是通过建立蓝牙连接，或者服务器和灯之间直接建立通信关系连接；服务器基于第一颜色序列，向终端发送控制指令，其中控制指令携带控制灯发出光线的颜色标识信息，终端设备接收到控制指令后，控制灯发射对应颜色的光线。或者，服务器可以直接向灯发送控制指令；灯接收到控制指令后，发射与控制指令中的颜色标识信息指示的颜色的光线。
另外，在目标设备发射光线时，针对每一种光线，可以持续一定的发光时长；在该发光时长内，与终端设备连接的图像获取设备获取目标人脸在该光线下的人脸图像。
在每种光线下获取的人脸图像可以是一张。例如基于第一颜色序列包括：红色、绿色以及蓝色；每种颜色持续的时长为2秒，且在2秒时间内，获取一张图像；具体过程为：控制目标设备发出红色光线持续2秒，并在发出红色光线的2秒内，获取1张与红色光线对应的第一人脸图像；然后控制目标设备发出绿色光线，在目标设备发出绿色光线的2秒内，获取1张与绿色光线对应的第一人脸图像；然后控制目标设备发出蓝色光线，在目标设备发出蓝色光线持续2秒，并在发出蓝色光线的2秒内，获取1张与蓝色光线对应的第一人脸图像。
每种光线下获取的人脸图像还可以是多张。例如基于第一颜色序列包括：红色、绿色以及蓝色；每种颜色持续的时长为5秒，且在5秒时间内，每秒获取一张图像，共获取与该颜色对应的5张图像；具体过程为：控制目标设备发出红色光线持续5秒，并在发出红色光线的5秒内，每秒获取1张与红色光线对应的第一人脸图像，共获取红色光线下的5张图像；然后控制目标设备发出绿色光线，在目标设备发出绿色光线的5秒内，每秒获取1张与绿色光线对应的第一人脸图像，共获取绿色光线下的5张图像；然后控制目标设备发出蓝色光线，在目标设备发出蓝色光线持续5秒，并在发出蓝色光线的5秒内，每秒获取1张与蓝色光线对应的第一人脸图像，共获取蓝色光线下的5张图像。
另外，为了保证获取的人脸图像中包括了目标人脸，提升检测的效率，在本公开另一实施例中，在控制目标设备发出与第一颜色序列中的多种颜色对应的光线时，包括：检测所述目标人脸是否位于所述目标设备发出光线后的光线覆盖区域内；响应于所述目标人脸位于所述光线覆盖区域内，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线。
其中，可以通过红外传感器检测目标人脸是否位于目标设备发出光线后覆盖的区域内。
示例性的，当检测到目标人脸位于光线覆盖区域内，将控制目标设备依次发出与第一颜色序列中的多种颜色分别对应的光线，从而能够获取与该颜色对应的第一人脸图像；当检测到目标人脸未位于光线覆盖区域内，目标设备不能依次发出与所述第一颜色序列中的多种颜色分别对应的光线，且不能从获取与该颜色对应的第一人脸图像，在该种情况下，当检测到目标人脸未位于光线覆盖区域内，可以提供语音提示用户位于光线覆盖区域内。
另外，也可以通过图像获取设备获取图像，并利用目标检测算法检测该图像中是否存在人脸；目标检测算法例如包括：基于神经网络构建的人脸检测算法。当检测到获取的图像中 存在人脸后，确定存光线覆盖区域内存在目标人脸。
针对上述S103：
在获取第一颜色序列中多种颜色下分别对应的第一人脸图像后，即可以对获取到每种颜色对应的第一人脸图像进行颜色检测、活体检测以及身份检测。
A：针对基于第一颜色序列，对第一人脸图像进行颜色检测的情况：
参见图2所示，本公开实施例提供了一种基于所述第一颜色序列，对所述第一人脸图像进行颜色检测的具体示例，包括：
S201：针对所述第一颜色序列中的每种颜色，对于该颜色对应的第一人脸图像进行颜色检测，得到与该颜色对应的第一人脸图像的颜色信息。
第一人脸图像是在拍摄人脸时获取的，同时对第一人脸图像中每一个像素点的颜色进行取样，在RGB色彩空间下，每个像素点的颜色有四个通道(R,G,B,Alpha)，基于第一人脸图像的每个像素点的(R,G,B,Alpha)得到与第一颜色序列中的每种颜色对应的第一人脸图像的颜色信息；颜色信息可以用来判断是否与第一颜色序列中的每种颜色对应。
S202：将所述每种颜色和对应第一人脸图像的颜色信息进行匹配。
将第一颜色序列中的每种颜色和对应的第一人脸图像的颜色信息进行匹配的方式有以下任一种：
(1)将获取到的第一人脸图像标记的颜色与目标设备发出的灯光颜色进行匹配。
基于第一颜色序列，在获取第一人脸图像时，目标设备发出的灯光颜色使得每张第一人脸图像标记了与灯光颜色对应的颜色，当每张第一人脸图像标记的颜色与目标设备发出的灯光颜色对应，则第一人脸图像与第一颜色序列中的每种颜色匹配成功。
(2)将获取到的第一人脸图像的时间戳与目标设备发出灯光的时间进行匹配。
目标设备发出每种灯光颜色的时间是预先设置好，而在目标设备发出每种灯光颜色的时间内将会获取第一人脸图像，例如，目标设备发出每种灯光颜色的时间持续5秒，则在这五秒内可以获取到5张人脸图像，即每张第一人脸图像的时间戳可以是1秒；将获取到第一人脸图像的时间戳与目标设备发出灯光的时间进行匹配，确定在目标设备发出每种灯光颜色的时间内获取到的多张第一人脸图像的颜色与对第一人脸图像进行颜色检测时的颜色进行匹配；若目标设备发出每种灯光颜色的时间内获取到的多张第一人脸图像的颜色与对第一人脸图像进行颜色检测时的颜色相同，则匹配成功。
S203：响应于所述第一颜色序列中的多种颜色分别与对应第一人脸图像的颜色信息匹配成功，确定所述第一人脸图像通过所述颜色检测。
基于第一颜色序列中的多种颜色，将与每种颜色对应的第一人脸图像的颜色信息进行匹配，若匹配成功，则第一人脸图像通过颜色检测。
B：针对基于第一颜色序列，对第一人脸图像进行活体检测的情况：
参见图3所示，本公开实施例提供了一种对第一人脸图像进行活体检测的具体示例，包括：
S301：针对所述第一颜色序列中的每种颜色，从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像。
在具体实施中，在确定与所述每种颜色对应的第一目标人脸图像时，包括：确定所述每种颜色对应的第一人脸图像的质量信息；基于所述质量信息，确定与所述每种颜色对应的第一目标人脸图像。
其中，第一人脸图像的质量信息例如包括以下至少一种：第一人脸图像的清晰度、第一 人脸图像中第一人脸的尺寸大小、第一人脸图像中第一人脸的完整度。
示例性的，每种颜色对应的第一人脸图像的清晰度达到预设阈值时，该第一人脸图像的质量评分较高，该评分表征第一人脸图像的质量的高低，可以随着清晰度大小发生改变；若该第一人脸图像的质量评分高，则该每种颜色对应的第一人脸图像为第一目标人脸图像；例如，系统第一人脸图像清晰度预设阈值为100，第一颜色序列中的红色对应的第一人脸图像的清晰度为100时，则该质量评分为100，则该红色对应的第一人脸图像为第一目标人脸图像；若第一颜色序列中的红色对应的第一人脸图像的清晰度为60时，则该质量评分为60，则该红色对应的第一人脸图像不是第一目标人脸图像。
第一人脸图像中第一人脸的尺寸大小的检测方式可以通过将第一人脸图像经过全卷积网络将得到特征图，特征图上每一个“点”对应该位置映射到第一人脸图像中属于第一人脸的区域，将第一人脸图像中的第一人脸区域视为人脸候选框，根据人脸候选框的大小，判断第一人脸图像中的第一人脸的尺寸大小。
由于用户的人脸尺寸大小是不变的，所以获取第一人脸图像时，将会计算用户的人脸尺寸大小，得到用户的预设人脸尺寸阈值。示例性的，每种颜色对应的第一人脸图像中的第一人脸的尺寸大小达到预设人脸尺寸阈值时，则该第一人脸图像的质量评分较高，该评分表征第一人脸图像的质量的高低；若该第一人脸图像的质量评分高，则该每种颜色对应的第一人脸图像为第一目标人脸图像；例如，系统第一人脸图像中的第一人脸尺寸大小预设阈值为150*150，第一颜色序列中的红色对应的第一人脸图像中的第一人脸的尺寸大小也为150*150时，则该质量评分可以为100，则该红色对应的第一人脸图像为第一目标人脸图像；若第一颜色序列中的红色对应的第一人脸图像中的第一人脸的尺寸大小为100*100时，则该质量评分为50，则该红色对应的第一人脸图像不是第一目标人脸图像。
第一人脸图像中第一人脸的完整度可以通过人脸关键点的数量来检测。基于人脸关键点检测模型对人脸关键点进行检测，得到人脸关键点的检测结果。例如人脸关键点能够检测到的人脸关键点有13个，而对某张人脸图像进行检测时，检测出的关键点有5个，则可以表征人脸不完整。根据人脸关键点的数量来判断第一人脸图像中的第一人脸的完整度；人脸关键点检测模型检测的人脸关键点数量、以及对第一人脸图像进行检测得到的实际数量之间的差值越大，则完整度越低。
例如，人脸图像中正脸的预设关键点数量为100，当获取到第一人脸图像后，对第一人脸图像中的第一人脸的正脸进行人脸关键点检测，当人脸关键点的数量大于90，则第一人脸图像中的第一人脸完整度为100分；当人脸关键点的数量在80～90之间，第一人脸图像中的第一人脸完整度为80分；当人脸关键点的数量在70～80之间，第一人脸图像中的第一人脸完整度为70分；当人脸关键点的数量在60～70之间，第一人脸图像中的第一人脸完整度为60分；第一人脸图像中的第一人脸的关键点越少评分越低。
示例性的，每种颜色对应的第一人脸图像中的第一人脸的完整度较高时，则该第一人脸图像的质量评分较高，该评分表征第一人脸图像的质量的高低，若该第一人脸图像的质量评分高，则该每种颜色对应的第一人脸图像为第一目标人脸图像；例如，第一人脸图像中的第一人脸完整度为100分，则该质量评分可以为100，则该红色对应的第一人脸图像为第一目标人脸图像；若第一颜色序列中的红色对应的第一人脸图像的完整度为60分时，则该质量评分为60，则该红色对应的第一人脸图像不是第一目标人脸图像。
S302：对所述每种颜色对应的第一目标人脸图像进行活体检测，得到所述每种颜色对应的第一目标人脸图像的活体检测结果。
具体实施中，根据第一颜色序列中的每种颜色都有一一对应的第一人脸图像，确定与每种颜色对应的第一目标人脸图像后，对第一目标人脸图像进行活体检测，根据对应的第一目标人脸图像的活体检测结果，判断该第一目标人脸图像是否为活体人脸。
对第一目标人脸图像可以使用光流场的方法进行活体检测，该方法主要利用图像序列中的像素强度数据的时域变化和相关性来确定各自像素位置的“运动”，从图像序列中得到各个像素点的运行信息，采用高斯差分滤波器、局部二值模式(Local Binary Patterns，LBP)纹理特征和支持向量机进行数据统计分析。同时，光流场对物体运动比较敏感，利用光流场可以统一检测眼球移动和眨眼；由于真实人脸并不是绝对静止的，存在微表情，如眼皮眼球的律动、眨眼、嘴唇及周边面颊的伸缩等，根据各个像素点的运行信息判断是否为活体人脸。
S303：响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测。在具体的实施方式中，确定所述第一人脸图像通过所述活体检测时，包括：响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量、且活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件，确定所述第一人脸图像通过所述活体检测。
活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，例如活体人脸预设数量为3张，当目标人脸为活体人脸的第一目标人脸图像为3张或者超过3张时，则该第一人脸图像符合达到预设数量的条件。
活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件，该预设条件例如可以包括以下b1～b3任一种：
b1：目标人脸为活体人脸的第一目标人脸图像质量评分最高。
当目标人脸为活体人脸的第一目标人脸图像在所有活体人脸的第一目标人脸图像中质量评分时最高的，则该目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件。
例如，所有目标人脸为活体人脸的第一目标人脸图像中质量评分有80分、75分、89分，则质量评分为89分的目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件；其他的目标人脸为活体人脸的第一目标人脸图像对应的质量信息不满足预设条件。
b2：目标人脸为活体人脸的第一目标人脸图像的质量评分大于预设的质量评分阈值。
例如，系统第一人脸图像预设的质量评分阈值为80分，若该目标人脸为活体人脸的第一目标人脸图像的质量评分为85分，则该目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件；若该目标人脸为活体人脸的第一目标人脸图像的质量评分为79分，则该目标人脸为活体人脸的第一目标人脸图像对应的质量信息不满足预设条件。
b3：目标人脸为活体人脸的第一目标人脸图像的质量评分大于预设的质量评分阈值的数量大于预设数量。
例如，系统第一人脸图像预设的质量评分阈值为80分，且活体人脸预设数量为3张；若目标人脸为活体人脸的第一目标人脸图像的质量评分为89分的数量有5张，则该目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件。
当目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量、且质量信息满足预设条件，则该第一目标人脸图像通过活体检测。
C：针对基于第一颜色序列，对第一人脸图像进行颜色检测的情况：
参见图4所示，本公开实施例提供一种基于第一人脸图像进行身份检测的具体示例，该步骤包括S401～S402，其中：
S401：基于所述第一颜色序列中每种颜色对应的第一人脸图像的质量信息，确定第二目 标人脸图像；
示例性的，将通过活体检测的第一人脸图像，基于第一颜色序列每种颜色对应的第一人脸图像的质量信息，选取质量信息较好的第一人脸图像确定为第二目标人脸图像，该确定方式与上述S301类似，在此不在赘述。
S402：基于所述第二目标人脸图像、以及预存的人脸登记图像进行身份检测。
预存的人脸登记图像是接收到人脸登记请求后存储在数据库中的人脸登记图像。
在基于第二目标人脸图像、以及预存的人脸登记图像进行身份检测时，例如可以先对第二目标人脸图像进行特征提取，得到第二目标人脸图像的第一特征信息；以及对预存的人脸登记图像进行特征提取，得到人脸登记图像的第二特征信息，然后确定第一特征信息和第二特征信息之间的距离。根据该距离，确定第二目标人脸图像、以及人脸登记图像是否属于同一用户。若两者属于同一用户，则通过身份检测。若属于不同用户，则未通过身份检测。
示例性的，将第二目标人脸图像与预存的人脸登记图像进行匹配，可以根据距离确定匹配评分。若匹配评分大于预设的匹配评分阈值，则该第二目标人脸图像确定与预存的人脸登记图像相同；例如，预设的匹配评分阈值为85，当第二目标人脸图像与预存的人脸登记图像匹配评分为90，则该第二目标人脸图像确定与预存的人脸登记图像相同即为同一人；当第二目标人脸图像与预存的人脸登记图像匹配评分为80，则该第二目标人脸图像确定与预存的人脸登记图像不相同即不是同一个人。
另外，本公开实施例还提供一种获得人脸登记图像的具体方法，参见图5所示，包括：
S501：响应于接收到人脸登记请求，获取第二颜色序列；所述第二颜色序列包括：多种颜色；
此处，第二颜色序列的获取方式和第一颜色序列的获取方式类似，具体可参见上述S101的详细说明，在此不再赘述。
S502：基于第二颜色序列，获取目标人脸在所述第二颜色序列中的多种颜色下分别对应的第二人脸图像。
此处，第二人脸图像的获取方式和第一人脸图像的获取方式类似，具体可参见上述S102的详细说明，在此不再赘述。
S503：基于所述第二颜色序列，对所述第二人脸图像进行颜色检测，以及，对所述第二人脸图像进行活体检测。
基于第二颜色序列，对第二人脸图像进行颜色检测、活体检测的方式，和上述基于第一颜色序列，对第一人脸图像进行颜色检测、以及活体监测的方式类似，具体方式参见上述S103的实施例，在此不在赘述。
S504：响应于所述第二人脸图像通过所述颜色检测、以及所述活体检测，基于所述第二人脸图像确定所述人脸登记图像。
当第二人脸图像通过颜色检测、活体检测后，可以根据第二人脸图像的图像质量，从第二人脸图像中确定人脸登记图像，并将人脸登记图像存入数据库中，用于进行身份检测。
针对上述S104：当第一人脸图像通过颜色检测、活体检测、以及身份检测后，将执行以下目标交互请求对应的交互操作D1～D3：
D1：目标交互请求包括：预约请求，执行与预约请求对应的交互操作，包括：
获取课程预约信息，并基于所述课程预约信息进行课程预约，生成课程预约记录。
示例性的，接收到预约请求后，可以执行课程预约操作，在获取到课程预约信息后，将会生成课程预约记录；该课程预约记录表示该用户可以在课程预约时间内进入驾校学习。
D2：目标交互请求包括：上课登记请求，执行与上课登记请求对应的交互操作，包括：
获取课程预约记录；
响应于所述课程预约记录指示预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道。
示例性的，获取课程预约记录即获取用户的预约时间以及预约课程，通过课程预约记录指示预约的课程开始时间是否满足预设时间条件，控制安检设备是否开启放行通道；其中课程预约记录包括了课程开始时间以及用户进行预约的当前时间。
预设时间条件可以包括以下(3)～(5)中任一种条件，其中：
(3)：预约的课程开始时间与当前时间处于同一天。
基于获取到的课程预约记录，确定预约的课程开始时间是否与当前时间是否为一天，若预约的课程开始时间与当前时间为一天，则预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道，用户可以进入驾校学习；例如预约的课程开始时间为9月1日下午2：00，用户在9月1日预约课程，该用户可以进入驾校学习。
(4)：预约的课程开始时间与当前时间之间的时间差小于预设时间差的阈值。
示例性的，基于获取到的课程预约记录，确定预约的课程开始时间与当前时间之间的时间差，若该预约的课程开始时间与当前时间之间的时间差小于预设时间差阈值，则预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道，用户可以进入驾校学习；例如，预约的课程开始时间为9月1日的14点，要求用户在9月1日8点至14点才可以进行课程预约即距离课程开始时间的6小时之内，用户在9月1日9点进行课程预约，可以进入驾校学习。
(5)：预约的课程开始时间位于当前时间之后，该预约的课程开始时间与当前时间之间的时间差小于预设时间差的阈值。
示例性的，基于获取到的课程预约记录，该预约的课程开始时间位于当前时间之后，确定预约的课程开始时间与当前时间之间的时间差，若该预约的课程开始时间与当前时间之间的时间差小于预设时间差阈值，则预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道，用户可以进入驾校学习；例如预约的课程开始时间为9月1日，要求用户需要在预设时间差阈值内进行课程预约即在距离课程开始时间的7天之内，才可以成功预约，用户在8月25日预约课程，则可以成功预约课程，在9月1日可以进入驾校学习。
(4)：预约的课程开始时间位于当前时间之前，该预约的课程开始时间与当前时间之间的时间差小于预设时间差的阈值。
示例性的，基于获取到的课程预约记录，该预约的课程开始时间位于当前时间之前，确定预约的课程开始时间与当前时间之间的时间差，若该预约的课程开始时间与当前时间之间的时间差小于预设时间差阈值，则预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道，用户可以进入驾校学习；例如预约的课程开始时间为9月1日，用户可以在课程开始之后的3天内预约课程，用户在9月2日预约课程，则可以成功预约课程进入驾校学习。
D3:目标交互请求包括：车辆实践操作请求，执行与车辆实践操作请求对应的交互操作，包括：
获取课程登记记录；
基于所述课程登记记录，控制与所述课程登记记录对应的用户信息进入排队状态；
响应于进入排队状态的用户信息达到排队结束条件，为所述用户信息分配用于车辆操作 实践的目标车辆。
示例性的，获取课程登记记录，可以控制课程登记记录对应的用户信息进入排队状态，当进入排队状态的用户信息达到排队结束条件，该用户可以进行车辆操作实践。
排队结束条件可以是用户预约车辆操作实践的时间到达的情况，用户可以进行车辆操作实践；例如A用户预约的车辆操作时间为下午的3点，则当天时间到达下午3点时，则A用户可以进行车辆操作学习。
排队结束的条件也可以是排在用户前面的其他用户车辆操作实践结束的情况，用户可以进行车辆操作实践；例如排在A用户前面的B用户的车辆操作实践结束后，A用户可以进行车辆操作实践。
本领域技术人员可以理解，在具体实施方式的上述方法中，各步骤的撰写顺序并不意味着严格的执行顺序而对实施过程构成任何限定，各步骤的具体执行顺序应当以其功能和可能的内在逻辑确定。
基于同一发明构思，本公开实施例中还提供了与交互控制方法对应的交互控制装置，由于本公开实施例中的装置解决问题的原理与本公开实施例上述交互控制方法相似，因此装置的实施可以参见方法的实施，重复之处不再赘述。
参照图6所示，为本公开实施例提供的一种交互控制装置的示意图，所述装置包括：获取模块61、检测模块62、处理模块63；其中，
获取模块61，用于响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；
检测模块62，用于基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；
处理模块63，用于响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。
一种可选的实施方式中，所述获取模块61，在所述基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像时，用于基于所述第一颜色序列，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，并在控制所述目标设备发射出每种颜色的光线时，获取与该种颜色对应的第一人脸图像。
一种可选的实施方式中，所述检测模块62，在所述控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线时，用于检测所述目标人脸是否位于所述目标设备发出光线后的光线覆盖区域内；响应于所述目标人脸位于所述光线覆盖区域内，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线。
一种可选的实施方式中，所述检测模块62，在基于所述第一颜色序列，对所述第一人脸图像进行颜色检测时，用于针对所述第一颜色序列中的每种颜色，对于该颜色对应的第一人脸图像进行颜色检测，得到与该颜色对应的第一人脸图像的颜色信息；将所述每种颜色和对应第一人脸图像的颜色信息进行匹配；响应于所述第一颜色序列中的多种颜色分别与对应第一人脸图像的颜色信息匹配成功，确定所述第一人脸图像通过所述颜色检测。
一种可选的实施方式中，所述检测模块62，在所述对所述第一人脸图像进行活体检测时，用于针对所述第一颜色序列中的每种颜色，从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像；对所述每种颜色对应的第一目标人脸图像进行活体检测，得到所述每种颜色对应的第一目标人脸图像的活体检测结果；响应于活体检测结果指示 所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测。
一种可选的实施方式中，还包括确定模块64，在所述从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像时，用于确定所述每种颜色对应的第一人脸图像的质量信息；基于所述质量信息，确定与所述每种颜色对应的第一目标人脸图像。
一种可选的实施方式中，所述确定模块64，在所述响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测时，用于响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量、且活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件，确定所述第一人脸图像通过所述活体检测。
一种可选的实施方式中，所述检测模块62，在所述基于所述第一人脸图像进行身份检测时，用于基于所述第一颜色序列中每种颜色对应的第一人脸图像的质量信息，确定第二目标人脸图像；基于所述第二目标人脸图像、以及预存的人脸登记图像进行身份检测。
一种可选的实施方式中，所述检测模块62，用于响应于接收到人脸登记请求，获取第二颜色序列；所述第二颜色序列包括：多种颜色；基于第二颜色序列，获取目标人脸在所述第二颜色序列中的多种颜色下分别对应的第二人脸图像；基于所述第二颜色序列，对所述第二人脸图像进行颜色检测，以及，对所述第二人脸图像进行活体检测；响应于所述第二人脸图像通过所述颜色检测、以及所述活体检测，基于所述第二人脸图像确定所述人脸登记图像。
一种可选的实施方式中，所述目标交互请求包括下述至少一种：
登录请求、课程购买请求、预约请求、上课登记请求、车辆实践操作请求。
一种可选的实施方式中，所述目标交互请求包括：预约请求，所述处理模块63，在所述执行与所述目标交互请求对应的交互操作时，用于获取课程预约信息，并基于所述课程预约信息进行课程预约，生成课程预约记录。
一种可选的实施方式中，所述目标交互请求包括：上课登记请求，所述处理模块63，在所述执行与所述目标交互请求对应的交互操作时，用于获取课程预约记录；响应于所述课程预约记录指示预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道。
一种可选的实施方式中，所述目标交互请求包括：车辆实践操作请求，所述处理模块63，在所述执行与所述目标交互请求对应的交互操作时，用于获取课程登记记录；基于所述课程登记记录，控制与所述课程登记记录对应的用户信息进入排队状态；响应于进入排队状态的用户信息达到排队结束条件，为所述用户信息分配用于车辆操作实践的目标车辆。
关于装置中的各模块的处理流程、以及各模块之间的交互流程的描述可以参照上述方法实施例中的相关说明，这里不再详述。
本公开实施例还提供了一种计算机设备，如图7所示，为本公开实施例提供的计算机设备结构示意图，包括：
处理器71和存储器72；所述存储器72存储有处理器71可执行的机器可读指令，处理器71用于执行存储器72中存储的机器可读指令，所述机器可读指令被处理器71执行时，处理器71执行下述步骤：
响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；
基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；
基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述 第一人脸图像进行身份检测；
响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。
上述存储器72包括内存721和外部存储器722；这里的内存721也称内存储器，用于暂时存放处理器71中的运算数据，以及与硬盘等外部存储器722交换的数据，处理器71通过内存721与外部存储器722进行数据交换。
上述指令的具体执行过程可以参考本公开实施例中所述的交互控制方法的步骤，此处不再赘述。
本公开实施例还提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被处理器运行时执行上述方法实施例中所述的交互控制方法的步骤。其中，该存储介质可以是易失性或非易失的计算机可读取存储介质。
本公开实施例还提供一种计算机程序产品，该计算机程序产品承载有程序代码，所述程序代码包括的指令可用于执行上述方法实施例中所述的交互控制方法的步骤，具体可参见上述方法实施例，在此不再赘述。
其中，上述计算机程序产品可以具体通过硬件、软件或其结合的方式实现。在一个可选实施例中，所述计算机程序产品具体体现为计算机存储介质，在另一个可选实施例中，计算机程序产品具体体现为软件产品，例如软件开发包(Software Development Kit，SDK)等等。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统和装置的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。在本公开所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本公开各个实施例中的各功能单元可以集成在一个处理单元中，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单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解，本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上所述实施例，仅为本公开的具体实施方式，用以说明本公开的技术方案，而非对其限制，本公开的保护范围并不局限于此，尽管参照前述实施例对本公开进行了详细的说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本公 开揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本公开实施例技术方案的精神和范围，都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应所述以权利要求的保护范围为准。

Claims (16)

  1. 一种交互控制方法，其特征在于，包括：
    响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；
    基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；
    基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；
    响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述身份检测，执行与所述目标交互请求对应的交互操作。
  2. 根据权利要求1所述的交互控制方法，其特征在于，所述基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像，包括：
    基于所述第一颜色序列，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，并在控制所述目标设备发射出每种颜色的光线时，获取与该种颜色对应的第一人脸图像。
  3. 根据权利要求2所述的交互控制方法，其特征在于，所述控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线，包括：
    检测所述目标人脸是否位于所述目标设备发出光线后的光线覆盖区域内；
    响应于所述目标人脸位于所述光线覆盖区域内，控制目标设备依次发出与所述第一颜色序列中的多种颜色分别对应的光线。
  4. 根据权利要求1-3任一项所述的交互控制方法，其特征在于，基于所述第一颜色序列，对所述第一人脸图像进行颜色检测，包括：
    针对所述第一颜色序列中的每种颜色，对于该颜色对应的第一人脸图像进行颜色检测，得到与该颜色对应的第一人脸图像的颜色信息；
    将所述每种颜色和对应第一人脸图像的颜色信息进行匹配；
    响应于所述第一颜色序列中的多种颜色分别与对应第一人脸图像的颜色信息匹配成功，确定所述第一人脸图像通过所述颜色检测。
  5. 根据权利要求1-4任一项所述的交互控制方法，其特征在于，所述对所述第一人脸图像进行活体检测，包括：
    针对所述第一颜色序列中的每种颜色，从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像；
    对所述每种颜色对应的第一目标人脸图像进行活体检测，得到所述每种颜色对应的第一目标人脸图像的活体检测结果；
    响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通过所述活体检测。
  6. 根据权利要求5所述的交互控制方法，其特征在于，所述从所述每种颜色对应第一人脸图像中，确定与所述每种颜色对应的第一目标人脸图像，包括：
    确定所述每种颜色对应的第一人脸图像的质量信息；
    基于所述质量信息，确定与所述每种颜色对应的第一目标人脸图像。
  7. 根据权利要求6所述的交互控制方法，其特征在于，所述响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量，确定所述第一人脸图像通 过所述活体检测，包括：
    响应于活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像的数量达到预设数量、且活体检测结果指示所述目标人脸为活体人脸的第一目标人脸图像对应的质量信息满足预设条件，确定所述第一人脸图像通过所述活体检测。
  8. 根据权利要求1-7任一项所述的交互控制方法，其特征在于，所述基于所述第一人脸图像进行身份检测，包括：
    基于所述第一颜色序列中每种颜色对应的第一人脸图像的质量信息，确定第二目标人脸图像；
    基于所述第二目标人脸图像、以及预存的人脸登记图像进行身份检测。
  9. 根据权利要求8所述的交互控制方法，其特征在于，还包括：响应于接收到人脸登记请求，获取第二颜色序列；所述第二颜色序列包括：多种颜色；
    基于第二颜色序列，获取目标人脸在所述第二颜色序列中的多种颜色下分别对应的第二人脸图像；
    基于所述第二颜色序列，对所述第二人脸图像进行颜色检测，以及，对所述第二人脸图像进行活体检测；
    响应于所述第二人脸图像通过所述颜色检测、以及所述活体检测，基于所述第二人脸图像确定所述人脸登记图像。
  10. 根据权利要求1-9任一项所述的交互控制方法，其特征在于，所述目标交互请求包括下述至少一种：
    登录请求、课程购买请求、预约请求、上课登记请求、车辆实践操作请求。
  11. 根据权利要求1-10任一项所述的交互控制方法，其特征在于，所述目标交互请求包括：预约请求，所述执行与所述目标交互请求对应的交互操作，包括：
    获取课程预约信息，并基于所述课程预约信息进行课程预约，生成课程预约记录。
  12. 根据权利要求1-11任一项所述的交互控制方法，其特征在于，所述目标交互请求包括：上课登记请求，所述执行与所述目标交互请求对应的交互操作，包括：
    获取课程预约记录；
    响应于所述课程预约记录指示预约的课程开始时间满足预设时间条件，控制安检设备开启放行通道。
  13. 根据权利要求1-12任一项所述的交互控制方法，其特征在于，所述目标交互请求包括：车辆实践操作请求，所述执行与所述目标交互请求对应的交互操作，包括：
    获取课程登记记录；
    基于所述课程登记记录，控制与所述课程登记记录对应的用户信息进入排队状态；
    响应于进入排队状态的用户信息达到排队结束条件，为所述用户信息分配用于车辆操作实践的目标车辆。
  14. 一种交互控制装置，其特征在于，包括：
    获取模块，用于响应于获取到目标交互请求，获取第一颜色序列；所述第一颜色序列包括：多种颜色；基于第一颜色序列，获取目标人脸在所述第一颜色序列中的多种颜色下分别对应的第一人脸图像；
    检测模块，用于基于所述第一颜色序列，对所述第一人脸图像进行颜色检测以及活体检测，并基于所述第一人脸图像进行身份检测；
    处理模块，用于响应于所述第一人脸图像通过所述颜色检测、所述活体检测、以及所述 身份检测，执行与所述目标交互请求对应的交互操作。
  15. 一种计算机设备，其特征在于，包括：处理器、存储器，所述存储器存储有所述处理器可执行的机器可读指令，所述处理器用于执行所述存储器中存储的机器可读指令，所述机器可读指令被所述处理器执行时，所述处理器执行如权利要求1至13任一项所述的交互控制方法的步骤。
  16. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被计算机设备运行时，所述计算机设备执行如权利要求1至13任一项所述的交互控制方法的步骤。
PCT/CN2022/125855 2022-03-24 2022-10-18 交互控制方法、装置、计算机设备及存储介质 Ceased WO2023178997A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210303610.3 2022-03-24
CN202210303610.3A CN114783018A (zh) 2022-03-24 2022-03-24 交互控制方法、装置、计算机设备及存储介质

Publications (1)

Publication Number Publication Date
WO2023178997A1 true WO2023178997A1 (zh) 2023-09-28

Family

ID=82424946

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/125855 Ceased WO2023178997A1 (zh) 2022-03-24 2022-10-18 交互控制方法、装置、计算机设备及存储介质

Country Status (2)

Country Link
CN (1) CN114783018A (zh)
WO (1) WO2023178997A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114783018A (zh) * 2022-03-24 2022-07-22 北京市商汤科技开发有限公司 交互控制方法、装置、计算机设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107832712A (zh) * 2017-11-13 2018-03-23 深圳前海微众银行股份有限公司 活体检测方法、装置和计算机可读存储介质
CN113408403A (zh) * 2018-09-10 2021-09-17 创新先进技术有限公司 活体检测方法、装置和计算机可读存储介质
CN114065980A (zh) * 2021-11-04 2022-02-18 支付宝(杭州)信息技术有限公司 自学车预约处理方法及装置
CN114783018A (zh) * 2022-03-24 2022-07-22 北京市商汤科技开发有限公司 交互控制方法、装置、计算机设备及存储介质

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113361349B (zh) * 2021-05-25 2023-08-04 北京百度网讯科技有限公司 人脸活体检测方法、装置、电子设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107832712A (zh) * 2017-11-13 2018-03-23 深圳前海微众银行股份有限公司 活体检测方法、装置和计算机可读存储介质
CN113408403A (zh) * 2018-09-10 2021-09-17 创新先进技术有限公司 活体检测方法、装置和计算机可读存储介质
CN114065980A (zh) * 2021-11-04 2022-02-18 支付宝(杭州)信息技术有限公司 自学车预约处理方法及装置
CN114783018A (zh) * 2022-03-24 2022-07-22 北京市商汤科技开发有限公司 交互控制方法、装置、计算机设备及存储介质

Also Published As

Publication number Publication date
CN114783018A (zh) 2022-07-22

Similar Documents

Publication Publication Date Title
US12536258B2 (en) Authenticating an identity of a person
CN110766033B (zh) 图像处理方法、装置、电子设备及存储介质
CN110119719B (zh) 活体检测方法、装置、设备及计算机可读存储介质
US20220083795A1 (en) Face living body detection method and apparatus, device, and storage medium
CN112528909B (zh) 活体检测方法、装置、电子设备及计算机可读存储介质
CN115147936B (zh) 一种活体检测方法、电子设备、存储介质及程序产品
CN108875546A (zh) 人脸身份验证方法、系统及存储介质
CN103383723A (zh) 用于生物特征验证的电子欺骗检测的方法和系统
CN102782718A (zh) 认证装置、认证程序以及认证方法
KR20220016529A (ko) 응시자 단말의 정면 카메라와 보조 카메라를 사용하여 부정행위를 방지하는 안면윤곽선 인식 인공지능을 사용한 온라인 시험 시스템 및 그 방법
CN114724071A (zh) 信息检测方法、装置、电子设备及存储介质
US12148247B2 (en) Information providing device, information providing method, and storage medium
CN114219661A (zh) 基于人工智能的理赔方法、装置、计算机设备和存储介质
CN118521936A (zh) 基于aigc检测算法识别ai换脸风险的方法及系统
KR20230017454A (ko) 비대면 평가에서의 부정행위 방지 방법, 장치 및 프로그램
CN113111810A (zh) 一种目标识别方法和系统
CN108171224A (zh) 一种电子相册及考勤机
WO2022222957A1 (zh) 一种目标识别的方法和系统
CN115171215A (zh) 一种无人抗原检测方法和装置
CN114783018A (zh) 交互控制方法、装置、计算机设备及存储介质
ES2947414T3 (es) Evaluación de la condición de los objetos del mundo real
CN109684945A (zh) 基于瞳孔的活体检测方法、装置、服务器及存储介质
JP2025019244A (ja) 判定装置、判定装置を含む携帯端末、および判定装置用のプログラム
CN114627518B (zh) 数据处理方法、装置、计算机可读存储介质和处理器
HK40069232A (zh) 交互控制方法、装置、计算机设备及存储介质

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: 22933048

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22933048

Country of ref document: EP

Kind code of ref document: A1

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 04/03/2025)

122 Ep: pct application non-entry in european phase

Ref document number: 22933048

Country of ref document: EP

Kind code of ref document: A1