Image acquisition method, electronic equipment, medium and device

Abstract

The invention discloses an image acquisition method, electronic equipment, a medium and a device. The method comprises the following steps: in a first image containing head image information of a target person, determining a second image without face image information of the target person, and when the number of the second image is determined to be greater than a preset image number, determining that the image information of the target person is greater than the preset image number; determining a second shooting device arranged in a second direction opposite to the installation direction of the first shooting device as a target shooting device, and when the image number of the second image is smaller than or equal to the preset image number, determining the deflection angle of the target person according to the face image information, and determining a target shooting device in a shooting device set based on the deflection angle, converting first coordinate information of the target person in the first image into second coordinate information of a third image shot by the target shooting device, and further extracting a front face image of the target person from the third image based on the second coordinate information. The accuracy and convenience of obtaining the front face image of the target person are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to an image acquisition method, an electronic device, a medium and an apparatus. BACKGROUND

[0002] In a remote video conference scenario, when a front face image of a target participant needs to be intercepted, the face image information of the target participant is usually intercepted from the captured image as the front face image for display through remote control or directly. Using remote control to obtain the front face information of the target participant involves manual participation and wastes human resources, and directly intercepting the face image of the target participant as the front face image has a certain deflection angle, resulting in inaccuracy of the intercepted front face image. It can be seen that the current acquisition of the front face information of the target participant in the video conference scenario has the problems of complicated acquisition process and inaccuracy. SUMMARY

[0003] The embodiments of the present application provide an image acquisition method, an electronic device, a medium and an apparatus, which can improve the accuracy and convenience of the acquired front face image of the target participant.

[0004] In a first aspect, the embodiments of the present application provide an image acquisition method, which comprises:

[0005] obtaining a plurality of first images captured by a first capturing device continuously, and extracting head image information of a target participant from each first image based on first coordinate information of the target participant in the first image;

[0006] determining a second image in which the face image information of the target participant does not exist in the head image information in the plurality of first images;

[0007] if the number of images of the second image is greater than a preset image number, determining a second capturing device facing the first capturing device as a target capturing device;

[0008] if the number of images of the second image is less than or equal to the preset image number, determining a deflection angle of the target participant relative to the first capturing device based on each face image information, and determining the target capturing device in a capturing device set based on a plurality of deflection angles;

[0009] converting the first coordinate information into second coordinate information of the target capturing device, and extracting a front face image of the target participant from a third image captured by the target capturing device based on the second coordinate information;

[0010] The first photographing device, the second photographing device, and the photographing device set are arranged in the same photographing space, the first photographing device is arranged at a first position in the photographing space, the second photographing device is arranged at a second position in the photographing space, the second position and the first position are relative positions in the photographing space, photographing devices in the photographing device set are arranged at a third position and a fourth position respectively, the third position and the fourth position are relative positions in the photographing space, and the first position and the second position are located on two sides of a first connection line between the third position and the fourth position.

[0011] In a second aspect, an embodiment of the present application provides a computer storage medium, which stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and performing steps of the method.

[0012] In a third aspect, an embodiment of the present application provides an electronic device, which comprises a processor and a memory, and the memory stores a computer program, and the computer program is suitable for being loaded by the processor and performing steps of the method.

[0013] In a fourth aspect, an embodiment of the present application provides a multi-camera cascade device, which comprises:

[0014] A head image extraction module is configured to acquire a plurality of first images obtained by continuous photographing of the first photographing device, and extract head image information of the target person from each first image based on first coordinate information of the target person in the first image.

[0015] An image determination module is configured to determine, in the plurality of first images, a second image in which face image information of the target person does not exist in the head image information.

[0016] A first photographing device determination module is configured to determine, if the number of images of the second image is greater than a preset image number, a second photographing device arranged opposite to the first photographing device as a target photographing device.

[0017] A second photographing device determination module is configured to determine, if the number of images of the second image is less than or equal to the preset image number, a deflection angle of the target person relative to the first photographing device based on each face image information, and determine the target photographing device in the photographing device set based on the plurality of deflection angles.

[0018] A front face information extraction module is configured to convert the first coordinate information into second coordinate information of a third image obtained by photographing of the target photographing device, and extract a front face image of the target person from the third image based on the second coordinate information.

[0019] The first shooting device, the second shooting device, and the set of shooting devices are arranged in the same shooting space, the first shooting device is arranged at a first position in the shooting space, the second shooting device is arranged at a second position in the shooting space, the second position and the first position are relative positions in the shooting space, the shooting devices in the set of shooting devices are arranged at a third position and a fourth position respectively, the third position and the fourth position are relative positions in the shooting space, and the first position and the second position are located on two sides of a first connection line between the third position and the fourth position.

[0020] In the embodiment of the present application, the target shooting device for obtaining the frontal face image of the target person can be determined in the case that the head image information of the target person in the first image obtained by shooting does not exist face image information, and the target shooting device for obtaining the frontal face image of the target person can be determined according to the deflection angle determined according to the face image information of the target person in the case that the face image information of the target person exists in the first image, so that the target shooting device for obtaining the frontal face image of the target person can be accurately determined, the first coordinate information of the target person in the first image is converted into second coordinate information of a third image shot by the target shooting device, and then the frontal face image of the target person is extracted from the third image based on the second coordinate information, so that the accuracy of the obtained frontal face image of the target person is improved, manual acquisition is not required, and the convenience of obtaining the frontal face image of the target person is improved. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1 A scene schematic diagram of an image acquisition method provided by an embodiment of the present application;

[0023] Figure 2 A flowchart schematic diagram of an image acquisition method provided by an embodiment of the present application;

[0024] Figure 3 An acquisition of face image information key point schematic diagram provided by an embodiment of the present application;

[0025] Figure 4 A flowchart schematic diagram of an image acquisition method provided by an embodiment of the present application;

[0026] Figure 5 A deflection angle schematic diagram provided by an embodiment of the present application;

[0027] Figure 6 A flowchart of an image acquisition method provided by an embodiment of the present application is shown in FIG. 1;

[0028] Figure 7 A structural diagram of a structure in which multiple photographing devices exist in a target orientation is shown in FIG. 2;

[0029] Figure 8 A flowchart of an image acquisition method provided by an embodiment of the present application is shown in FIG. 1;

[0030] Figure 9 A structural diagram of an information display device provided by an embodiment of the present application is shown in FIG. 3;

[0031] Figure 10 A structural diagram of a multi-cascaded device provided by an embodiment of the present application is shown in FIG. 4;

[0032] Figure 11 A structural diagram of a multi-cascaded device provided by an embodiment of the present application is shown in FIG. 4;

[0033] Figure 12 A structural diagram of a multi-cascaded device provided by an embodiment of the present application is shown in FIG. 4;

[0034] Figure 13 A structural diagram of a multi-cascaded device provided by an embodiment of the present application is shown in FIG. 4. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without any creative work fall within the scope of protection of the present application.

[0036] In the description of the present application, it is understood that the terms "first", "second" and the like are only for the purpose of description and cannot be understood as indicating or implying relative importance. In the description of the present application, it is pointed out that, unless otherwise explicitly specified and limited, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, the process, method, system, product or equipment including a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units not listed, or optionally also includes other steps or units inherent to these processes, methods, products or equipment. The specific meaning of the above terms in the present application can be understood by the person skilled in the art according to the specific circumstances. In addition, in the description of the present application, "multiple" means two or more, unless otherwise specified. The association relationship of the associated objects is described, which means that there can be three relationships, for example, A and / or B can represent three cases: A exists alone, A and B exist together, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after it.

[0037] In the prior art, when the front face image of the target person participating in the conference needs to be intercepted, the face image information of the target person is usually intercepted as the front face image for display through remote control or directly from the captured image. Using remote control to obtain the front face information of the target person has the problem of requiring manual participation and wasting human resources, and directly intercepting the face image of the target person as the front face image has a certain deflection angle, resulting in inaccuracy of the intercepted front face image. Therefore, the current acquisition of the front face information of the target person in the video conference scene has the problems of complicated acquisition process and inaccuracy.

[0038] In the embodiments of the present application, by determining the target shooting device for acquiring the front face image of the target person in the case that the head image information of the target person in the first image obtained by shooting does not exist face image information, and determining the target shooting device for acquiring the front face image of the target person according to the deflection angle determined by the face image information in the case that the face image information of the target person exists in the first image, the target shooting device for acquiring the front face image of the target person can be accurately determined, and by converting the first coordinate information of the target person in the first image into second coordinate information of a third image shot by the target shooting device, and then extracting the front face image of the target person from the third image based on the second coordinate information, the accuracy of the acquired front face image of the target person is improved, and manual acquisition is not required, improving the convenience of acquiring the front face image of the target person.

[0039] The multi-camera cascade device can adopt at least two cameras to collect image information in a shooting space. The multi-camera cascade method provided in the embodiments of the present application can be realized by relying on a computer program and can run on a multi-camera cascade device based on the von Neumann system. The computer program can be integrated in an application or can run as an independent tool application. Referring to Figure 1 , Figure 1 a scene diagram for use of the embodiments of the present application.

[0040] The multi-camera cascade device obtains a plurality of first images of a conference room (shooting space) obtained by continuous shooting of a first camera, extracts head image information of a target person A from each first image, and determines a second image in which the face image information of the target person A does not exist in the head image information among the plurality of first images.

[0041] In one case, if the number of images of the second image is greater than a preset number of images, a second camera disposed opposite to the first camera is determined as a target camera, and a front face image of the target person A is obtained by shooting through the target camera.

[0042] In another case, the number of images of the second image is less than or equal to the preset number of images, the deflection angle of the target person A relative to the first camera is determined based on each face image information, the target camera is determined in the camera set based on the plurality of deflection angles, and the front face image of the target person A is extracted from a third image obtained by the target camera through the coordinate information of the head image information in the first image.

[0043] Optionally, after the front face image of the target person A is obtained, the front face image of the target person A can be displayed in a preset display area of the multi-camera cascade device when the target person A speaks, so that the participants can quickly check the target person A who is currently speaking.

[0044] In the embodiments of the present application, the target camera for obtaining the front face image of the target person can be determined in the case that the head image information of the target person in the obtained first image does not exist face image information, and the target camera for obtaining the front face image of the target person can be determined according to the deflection angle determined based on the face image information in the case that the face image information of the target person exists in the first image. The target camera for obtaining the front face image of the target person can be accurately determined, the front face image of the target person is obtained through the target camera, and the accuracy of the obtained front face image of the target person is improved. At the same time, manual acquisition is not required, and the convenience of obtaining the front face image of the target person is improved.

[0045] The image acquisition method provided in the embodiments of the present application will be described in detail below. Figures 2-9

[0046] ​Please refer to Figure 2 A flowchart of an image acquisition method is provided for the embodiments of the present application. As shown in Figure 2 The image acquisition method can include the following steps S101-S105.

[0047] S101, obtaining a plurality of first images obtained by a first photographing device continuously photographing, and extracting head image information of a target person from each first image based on first coordinate information of the target person in the first image;

[0048] The first photographing device can be a device for photographing all participants in a photographing space, which can be a wide-angle photographing device, thereby widening the range of photographing and obtaining more first images of participants in the photographing space.

[0049] The first photographing device is installed at a first position of the photographing space. Optionally, in an embodiment, the first position can be the installation position of the electronic device containing the multi-stage device. Based on this, the first photographing device can be installed on the display screen of the electronic device.

[0050] The first coordinate information can be coordinate information representing the target person in the first image; and the head image information can be image information containing all organs above the cervical spine of the target person.

[0051] Optionally, after the multi-stage device obtains the first image, when extracting the head image information of the target person, the edge profile of the person present in the first image can be extracted based on the edge detection operator sobel, and then the edge profile of the target person is determined in the edge profile of the first image according to the first coordinate information, the head profile of the target person is selected from the edge profile of the target person according to the shape characteristics of the edge profile, and then the region image of the region where the head profile is located is intercepted to obtain the head image information of the target person. Then, non-maximum suppression is applied to filter non-edge pixel points in the head image information, the head image information is further refined, and finally a double-threshold edge selection method is used to obtain accurate head image information of the target person. In the embodiments of the present application, after the head image information of the target person is intercepted based on the edge detection operator sobel, the intercepted head image information of the target person can be directly used as the head image information of the target person extracted from the first image, and the head image information of the target person obtained after filtering can also be used as the head image information of the target person extracted from the first image, thereby improving the quality of the head image information of the target person.

[0052] It can be understood that in the conference scene, since the positions of the conference participants are fixed, after the first shooting device obtains the first coordinate information of the target person in any first image, the head image information of the target person can be extracted in the plurality of first images based on the first coordinate information. Thus, the multi-cascade device does not need to determine the first coordinate information of the target person in each first image, and the head image information of the target person can be extracted from the plurality of first images, saving computing power, wherein the first coordinate information can be composed of the contour information of the target person in the first image.

[0053] In one case, the average of the coordinate information of the contour information of the target person in the plurality of first images can also be used as the first coordinate information.

[0054] In another case, the coordinate information of the contour information of the target person in the plurality of first images can be obtained, and the set formed by the plurality of coordinate information is regarded as the first coordinate information.

[0055] Optionally, the multi-camera cascade device can be applied in a conference scene, and a user can specify to display a close-up image of a conference participant who is speaking, so the multi-camera cascade device can confirm the person who is speaking in the target scene as the target person. After obtaining the first image, the multi-camera cascade device can obtain the mouth opening area of each person in the first image, and if the mouth opening area of a person in the first image reaches a preset area, the person is determined to be the target person, wherein the target person can be one or more, and the preset area can be an initial setting of the multi-camera cascade device, or can be set by a user or relevant staff, for example: the preset area can be 5 square centimeters.

[0056] It can be understood that since the distance between the camera and the person, the focal length of the camera, and the like are different, the multi-camera cascade device can have difficulty in obtaining the specific area size of the mouth opening area of the person, so the multi-camera cascade device can obtain the percentage of the mouth opening area of the person to the face area of the person, and if the percentage is greater than a preset percentage, the person can be confirmed as the target person, wherein the preset percentage can be an initial setting of the multi-camera cascade device, or can be set by a user or relevant staff, and the preset percentage can be 5%.

[0057] Optionally, the multi-cascade device can also determine the position information of the conference participant who is currently speaking through a microphone, determine the relative position of the position information and the first shooting device, and then determine the target person in the plurality of first images based on the relative position.

[0058] Optionally, when the multi-cascade device needs to perform a face close-up on all conference participants, it regards all conference participants in the first image as target persons.

[0059] Optionally, when the multi-cascade device needs to take a close-up of the face of the specified participant, the face image information of the target person needs to be entered in advance, and then after obtaining a plurality of first images, the first images are subjected to face recognition, and the participant whose face image information in the first image matches the pre-entered face image information is determined as the target person.

[0060] S102, determining, in the plurality of first images, a second image in which the face image information of the target person does not exist in the head image information;

[0061] The second image can be the head image information of the target person extracted in the first image, but the image in which the face image information of the target person does not exist in the head image information.

[0062] After the multi-cascade device extracts the head image information in the first image, it further determines a second image in which the face image information of the target person does not exist in the extracted head image information.

[0063] Optionally, after the multi-cascade device obtains the head image information, it uses a face detection algorithm (for example, Viola-Jones algorithm) to determine whether the face image information of the target person exists in the head image information, and if it is determined that the face image information of the target person exists in the head image information, it further extracts the face image information of the target person; if it is detected that the face image information of the target person does not exist in the head image information, the first image containing the head image information of the target person is determined as the second image.

[0064] S103, if the number of images of the second image is greater than a preset number of images, determining a second shooting device arranged opposite to the first shooting device as a target shooting device;

[0065] The preset number of images can be a number set for determining the target shooting device from the set of shooting devices and the second shooting device, and the value of the number can be determined according to the number of obtained first images, for example, when the number of first images is 6, the value can be set to 3.

[0066] When the multi-cascade device determines that the number of images of the second image is greater than the preset number of images, it determines the second shooting device as the target shooting device, and obtains the front face image of the target person from the images obtained by the second shooting device. The second shooting device is a device installed in a second orientation in the shooting space, and the second orientation in which it is installed is opposite to the first orientation in which the first shooting device is located in the shooting space.

[0067] Specifically, when the multi-cascade device determines that the number of images of the second image is greater than the preset number of images, it determines that the face of the target person is deflected to the second orientation, and then determines the second shooting device installed in the second orientation as the target shooting device.

[0068] For example, when the multi-cascade device determines the six acquired first images, it extracts the head information of the target person in each first image. When it is determined that the number of second images in which the facial image information of the target person does not exist in the head image information of the target person in the first image is four, which is greater than the preset number of images, it is determined that the face of the target person is deflected toward the second direction, and the second shooting device installed in the second direction is determined to be the target shooting device.

[0069] S104, if the number of the second images is less than or equal to the preset number of images, determining a deflection angle of the target person relative to the first camera based on information of each facial image, and determining a target camera in the camera set based on the multiple deflection angles;

[0070] When the multi-cascade device determines that the number of second images is less than or equal to the preset number of images, it further determines a deflection angle of the target person relative to the first shooting device using facial image information extracted from each of the plurality of first images, thereby obtaining a plurality of deflection angles. The target shooting device is then determined from the set of shooting devices based on the plurality of deflection angles.

[0071] The deflection angle may be a deflection size of the target person's face relative to the first camera when the face of the target person is deflected.

[0072] After acquiring the facial image information of the target person, the multi-cascade device can perform facial angle detection on the facial image information, and then obtain the deflection angle of the target person's face relative to the first shooting device.

[0073] Reference Figure 3 As shown, Figure 3 Schematic diagram of facial angle detection for obtaining facial image information. Figure 3 After obtaining facial image information of a target person, the multi-cascade device determines key point information for facial angle detection within the facial image information, for example, by using a 3DMM (3D Morphable Model) algorithm to perform facial angle detection on the facial image information. In this embodiment, image information such as eyebrows, eyes, nose, mouth, and jawline contained in the facial information can be extracted as key point information for facial angle detection, thereby determining the deflection angle of the target person relative to the first camera device.

[0074] The shooting device set may be a set of shooting devices including a first orientation, a third orientation, and a fourth orientation, wherein the shooting devices in the shooting device set are all arranged in the same shooting space, wherein the third orientation and the fourth orientation are relative orientations in the shooting space, the second orientation and the first orientation are relative orientations in the shooting space, and the first orientation and the second orientation are located on both sides of the first connecting line between the third orientation and the fourth orientation.

[0075] Specifically, the multi-cascade device determines that the face of the target person is deflected to one of the first orientation, the third orientation and the fourth orientation, determines the deflection angle of the target person relative to the first photographing device, and determines the target photographing device from the set of photographing devices based on the plurality of deflection angles when the number of images of the second image is less than or equal to the preset number of images.

[0076] For example, the multi-cascade device extracts head information of the target person in each of the first images when it is determined that six first images are obtained, and determines that the face of the target person is deflected to one of the first orientation, the third orientation and the fourth orientation, and determines the deflection direction of the face of the target person relative to the first photographing device when the number of images of the second image in which the head image information of the target person does not exist in the first image of the target person is two, which is less than the preset number of images. The target photographing device is determined from the set of photographing devices according to the deflection angle.

[0077] S105, converting the first coordinate information into second coordinate information of a third image obtained by the target photographing device, and extracting a front face image of the target person from the third image based on the second coordinate information;

[0078] The third coordinate information can be information representing the position of the target person in the third image.

[0079] After the multi-cascade device determines the first coordinate information of the head image information in the first image, the first coordinate information in the image coordinate system of the first photographing device is converted into coordinate information in the world coordinate system according to the internal and external parameters of the first photographing device, and then the coordinate information in the world coordinate system is converted into second coordinate information in the image coordinate system of the target photographing device according to the internal and external parameters of the target photographing device. Thus, the front face image of the target person can be extracted from the third image obtained by the target photographing device according to the finally obtained second coordinate information.

[0080] In the embodiments of the present application, the target photographing device for obtaining the front face image of the target person can be determined in the case that the head image information of the target person in the obtained first image does not exist face image information, and the target photographing device for obtaining the front face image of the target person can be determined according to the deflection angle determined based on the face image information of the target person in the case that the face image information of the target person exists in the first image. The target photographing device for obtaining the front face image of the target person can be accurately determined, the front face image of the target person is obtained through the target photographing device, and the accuracy of the obtained front face image of the target person is improved. At the same time, manual operation is not required, and the convenience of obtaining the front face image of the target person is improved.

[0081] Reference Figure 4, Figure 4 This is a flow chart of an embodiment of the image acquisition method of this application. Figure 4 As shown, the image acquisition method may include the following steps S201 to S203.

[0082] S201, determining an average angle value of a plurality of deflection angles and a variance value of the plurality of deflection angles;

[0083] After obtaining the multiple deflection angles, the multi-cascade device calculates the average value and variance value of the multiple deflection angles.

[0084] Reference Figure 5 , Figure 5 This is a schematic diagram of the deflection angle of this application. Figure 5 As shown, in this embodiment, the deflection angle is based on the target person facing the first shooting device as a reference point. If the deflection angle rotates clockwise toward the fourth direction, the deflection angle is a negative number; if the deflection angle rotates counterclockwise toward the third direction, the deflection angle is a positive number.

[0085] The multi-cascade device determines the deflection angle of the target person relative to the first camera device based on each facial image of the target person, thereby obtaining multiple deflection angles of the target person based on the multiple facial images of the target person, and calculating the average angle value and variance value of the multiple deflection angles.

[0086] The formula for calculating the average angle value is: in is the average angle value, X n is the nth deflection angle, and n is the total number of deflection angles.

[0087] The formula for calculating the variance value is: Where s is the variance value.

[0088] S202, determining a target orientation corresponding to a facial deflection direction of the target person based on the average angle value and the variance value;

[0089] The facial deflection direction may be a direction in which the target person's face is deflected relative to the first shooting device.

[0090] The target direction may be the direction toward which the face of the target person is turned.

[0091] After obtaining the average angle value and the variance value, the multi-cascade device determines the target orientation corresponding to the facial deflection direction of the target person based on the average angle value and the variance value.

[0092] In this embodiment, the multi-cascade device determines the target orientation corresponding to the facial deflection direction of the target person based on the average angle value and the variance value, which can be determined specifically according to the relationship between the average angle value and the preset angle range, and the size between the variance value and the preset variance value.

[0093] The preset angle range can be a set value for determining whether the average angle is in the preset angle range interval, and the value can be determined according to the installation position of the first shooting device and the size of the shooting angle. For example, if the first shooting device can shoot a front face image of a person within a 10° deflection angle range in the third and fourth directions with the vertical direction as the origin, the preset angle range can be determined as [-10°, 10°].

[0094] The preset variance value can be a preset value for determining whether the obtained multiple deflection angles meet the angle change range. The value can be determined according to the preset angle range.

[0095] Specifically, after the multi-cascade device obtains the average angle value and the variance value of the multiple deflection angles, when it is determined that the average angle value is included in the preset angle range and the variance value is less than the preset variance threshold, the target direction corresponding to the face deflection direction of the target person is determined as the first direction; when it is determined that the average angle value is not included in the preset angle range or the variance value is greater than or equal to the preset variance value, the target direction corresponding to the face deflection direction of the target person is determined based on the multiple deflection angles.

[0096] For example, when the multi-cascade device determines the deflection angles of the target person relative to the first shooting device to be 10°, -8°, 5°, and -3° through face image information, the average angle value is calculated to be 1° and the variance value is calculated to be 48.5, the average angle value is included in the preset angle range, and the variance value is less than the preset variance value. Then, the target direction corresponding to the face deflection direction of the target person is determined as the first direction. When the multi-cascade device determines the deflection angles of the target person relative to the first shooting device to be -18°, -11°, -13°, and 20° through face image information, the average angle value is calculated to be -5.5° and the variance value is calculated to be 223.25, the average angle value is included in the preset angle range, but the variance value is greater than the preset variance value. Then, the target direction corresponding to the face deflection direction of the target person is determined based on the multiple deflection angles.

[0097] It can be understood that when the average angle value of the target person relative to the first shooting device is not included in the preset angle range or the variance value is greater than or equal to the preset variance value, it indicates that the deflection angle of the target person relative to the first shooting device is greater than the shooting range of the first shooting device, and the target direction corresponding to the face deflection direction of the target person needs to be further determined based on the deflection angle.

[0098] Specifically, a first deflection angle greater than a first angle threshold is determined from the plurality of deflection angles, a first angle number of the first deflection angle is determined, when the first angle number is greater than or equal to a first preset angle number, a target direction corresponding to the deflection direction of the face of the target person is determined as a third direction. A second deflection angle less than a second angle threshold is determined from the plurality of deflection angles, a second angle number of the second deflection angle is determined, when the second angle number is greater than or equal to the first preset angle number, the target direction corresponding to the deflection direction of the face of the target person is determined as a fourth direction.

[0099] The first angle threshold can be a maximum angle value at which the first shooting device can obtain the front face information of the target person, and the first shooting device is taken as the origin and rotates counterclockwise from the vertical direction to the third direction.

[0100] The second angle threshold can be a maximum angle value at which the first shooting device can obtain the front face information of the target person, and the first shooting device is taken as the origin and rotates counterclockwise from the vertical direction to the fourth direction.

[0101] The first preset angle number can be an angle number value set for determining the target direction corresponding to the deflection direction of the face of the target person, and the value can be determined according to the image number of the first image. For example, when the image number of the first image is 4, the first preset angle number can be determined as 2.

[0102] For example, when the first angle threshold is 10°, the second angle threshold is -10°, and the first preset angle number is 2, the multi-cascading device determines that the deflection angles of the target person relative to the first shooting device are -18°, -11°, -13°, and 20° through the face image information, calculates an average angle value of -5.5° and a variance value of 223.25, and needs to determine the target direction corresponding to the deflection direction of the face of the target person through the deflection angle. The multi-cascading device determines that the first deflection angle greater than the first angle threshold is 20°, the first angle number of the first deflection angle is 1, which is less than the first preset angle number, and determines that the target direction corresponding to the deflection direction of the face of the target person is the fourth direction; or the multi-cascading device determines that the second deflection angle less than the second angle threshold is -18°, -11°, and -13°, the second angle number of the second deflection angle is 3, which is greater than the first preset angle number, and determines that the target direction corresponding to the deflection direction of the face of the target person is the fourth direction.

[0103] It can be understood that in the embodiment, the manner in which the first angle number is greater than or equal to the first preset angle number can also be as described above, and will not be described again here.

[0104] In the embodiment of the present application, the target direction corresponding to the face deflection direction of the target person is determined based on a plurality of deflection angles, the target direction in which the target person deflects multiple times is determined as the target direction, and then the target shooting device is determined in the target direction, thereby improving the accuracy of determining the target shooting device.

[0105] In the embodiment of the present application, the target direction corresponding to the face deflection direction of the target person is determined based on the average angle value and the preset angle range, and the size of the variance value and the preset variance value, which can more accurately determine the target direction.

[0106] In S203, the target shooting device is determined in the shooting device set based on the target direction.

[0107] After the target direction is determined by the multi-cascading device, the target shooting device is determined in the shooting device set based on the target direction.

[0108] Specifically, when the target direction is determined as the first direction, the multi-cascading device determines the first shooting device installed in the first direction as the target shooting device for obtaining the frontal image of the target person.

[0109] For example, when the target direction is determined as the first direction, the multi-cascading device determines the first shooting device installed in the first direction as the target shooting device, and calls the target shooting device to obtain the frontal image of the target person. When obtaining the frontal image of the target person, the multi-cascading device can obtain an image corresponding to the time range of the timestamp of the first image as a third image in the image obtained by the target shooting device, and then extract the frontal image of the target person based on the third image.

[0110] In the embodiment of the present application, when the target direction corresponding to the face deflection direction of the target person is the first direction, the first shooting device installed in the first direction is determined as the target shooting device, which realizes that when the target direction corresponding to the face deflection direction of the target person is the first direction, the frontal image of the target person is accurately obtained by the first shooting device.

[0111] Specifically, when the target direction is determined as the third direction, the multi-cascading device determines the third shooting device installed in the third direction as the target shooting device for obtaining the frontal image of the target person.

[0112] For example, when the target direction is determined as the third direction, the multi-cascading device determines the third shooting device installed in the third direction as the target shooting device, and calls the target shooting device to obtain the frontal image of the target person. When obtaining the frontal image of the target person, the multi-cascading device can obtain an image corresponding to the time range of the timestamp of the first image as a third image in the image obtained by the target shooting device, and then extract the frontal image of the target person based on the third image.

[0113] Specifically, the multi-cascade device determines the fourth photographing device at the fourth orientation as the target photographing device for obtaining the frontal image of the target person when the target orientation is the fourth orientation.

[0114] For example, when the target orientation is the fourth orientation, the multi-cascade device determines the fourth photographing device installed at the fourth orientation as the target photographing device, and calls the target photographing device to obtain the frontal image of the target person. When obtaining the frontal image of the target person, the multi-cascade device can obtain an image corresponding to the time range of the timestamp of the first image as a third image from the image obtained by the target photographing device, and further extract the frontal image of the target person based on the third image.

[0115] In the embodiments of the present application, when the target orientation corresponding to the face deflection direction of the target person is the third orientation or the fourth orientation, the third photographing device installed at the third orientation or the fourth photographing device installed at the fourth orientation is determined as the target photographing device, so that when the target orientation corresponding to the face deflection direction of the target person is the third orientation or the fourth orientation, the frontal image of the target person can be accurately obtained by the target photographing device.

[0116] In the embodiments, the target orientation corresponding to the face deflection direction of the target person is determined based on the change amount of the plurality of deflection angles, and then the target photographing device is obtained, so that when the deflection direction of the target person changes, the face information of the target person can also be obtained.

[0117] Referring to Figure 6 , Figure 6 FIG. 1 is a flowchart of an embodiment of an image acquisition method of the present application. As shown in FIG. 1, the image acquisition method can include the following steps S301-S303. Figure 6

[0118] S301, if there are a plurality of photographing devices in the target orientation, determine the angle intervals in which the plurality of deflection angles are respectively located, and each angle interval corresponds to one photographing device in the plurality of photographing devices;

[0119] As shown in FIG. 2, Figure 7 FIG. 2 is a structural diagram of the case where there are a plurality of photographing devices in the target orientation. Figure 7

[0120] In FIG. 2, there are photographing device A, photographing device B, photographing device C, and photographing device D in the third orientation, and each photographing device is installed at the third orientation of the photographing space according to a preset distance (1 meter). Figure 7 When the multi-cascade device determines that there are a plurality of photographing devices in the target orientation (the third orientation), the angle intervals in which the plurality of deflection angles are located are determined.

[0121]

[0122] ​​​The angle interval is a preset interval, and the photographing device in different angle intervals can obtain the front face image of the target person at different angles of the target direction.

[0123] For example, the angle interval corresponding to the photographing device A is (10°, 30°], the angle interval corresponding to the photographing device B is (30°, 50°], the angle interval corresponding to the photographing device C is (50°, 70°], and the angle interval corresponding to the photographing device D is (70°, 90°].

[0124] The multi-cascade device obtains the deflection angles of the target person relative to the first photographing device as 10°, 35°, 37°, 38°, and 40°, determines that the angle interval corresponding to the deflection angle 10° is (10°, 30°], and determines that the angle interval corresponding to the deflection angles 35°, 37°, 38°, and 40° is (30°, 50°].

[0125] S302, determine the third angle quantity of the deflection angles included in each angle interval;

[0126] The multi-cascade device further determines the third angle quantity included in each angle interval, wherein the third angle quantity included in the angle interval (10°, 30°] is 1, and the third angle quantity included in the angle interval (30°, 50°] is 4.

[0127] S303, determine the target photographing device in the plurality of photographing devices at the target direction based on the third angle quantity.

[0128] Specifically, after the multi-cascade device determines the third angle quantity of the deflection angles included in the angle interval, the target angle interval is determined according to the third angle quantity, and the photographing device corresponding to the target angle interval is determined as the target photographing device.

[0129] For example, the multi-cascade device determines that the third angle quantity included in the angle interval (10°, 30°] is 1, and the third angle quantity included in the angle interval (30°, 50°] is 4, determines that the interval including the largest third angle quantity is the target photographing interval, that is, determines that the angle interval (30°, 50°] is the target angle interval, and further determines that the photographing device B corresponding to the target angle interval is the target photographing device. By determining the angle interval including the largest third angle quantity as the target angle interval, the target angle interval is determined according to the plurality of deflection angles, and the target photographing device is determined, thereby improving the accuracy of determining the target photographing device.

[0130] In the embodiment of the present application, multiple shooting devices are arranged in the target direction, and then the target shooting device is determined according to multiple deflection angles of the target person. The front face image of the target person in different angle intervals of the target direction is obtained by different shooting devices, and a more accurate front face image of the target person can be obtained.

[0131] With reference to Figure 8 , Figure 8 FIG. 1 is a schematic diagram of an embodiment of an image acquisition method of the present application. As shown in FIG. 1, the image acquisition method can include the following steps S401-S403. Figure 8

[0132] S401, if multiple second shooting devices exist, multiple target images are acquired based on any second shooting device;

[0133] S402, target face image information of the target person in each target image is extracted, and a target deflection angle of the target person relative to the second shooting device is determined based on the target face image information;

[0134] The target image can be an image of the whole personnel obtained by the second shooting device.

[0135] The target face image information can be face image information obtained from the target face image.

[0136] The target deflection angle can be a face deflection angle determined based on the target face image information.

[0137] In the embodiment, when it is determined that multiple second shooting devices exist, multiple target images are acquired based on any second shooting device, target face image information of the target person in each target image is extracted, and a target deflection angle of the target person relative to the second shooting device is determined based on the target face image information.

[0138] It can be understood that the process of determining the target deflection angle of the target person relative to the second shooting device in the embodiment is similar to the determination of the target deflection angle of the target person relative to the first shooting device, which will not be described here.

[0139] S403, angle intervals in which the multiple target deflection angles are located are determined, and each angle interval corresponds to one of the multiple shooting devices;

[0140] The angle interval is a pre-set interval, and the shooting device in different angle intervals can obtain the front face image of the target person deflected at different angles of the target direction.

[0141] For example, in the second direction, there are a shooting device E, a shooting device F, a shooting device G, and a shooting device H, and each shooting device is installed in the second direction of the shooting space according to a preset distance (1 meter).​

[0142] The angle interval corresponding to the photographing device E is (10°, 30°], the angle interval corresponding to the photographing device F is (30°, 50°], the angle interval corresponding to the photographing device G is (50°, 70°], and the angle interval corresponding to the photographing device H is (70°, 90°].

[0143] The multi-cascade device obtains the deflection angles of the target person relative to the second photographing device as 10°, 35°, 37°, 38° and 40°, determines that the angle interval corresponding to the deflection angle 10° is (10°, 30°], and determines that the angle interval corresponding to the deflection angles 35°, 37°, 38° and 40° is (30°, 50°].

[0144] S404, determining a fourth angle quantity of the target deflection angles included in each angle interval;

[0145] The multi-cascade device further determines the fourth angle quantity included in each angle interval, wherein the third angle quantity included in the angle interval (10°, 30°] is 1, and the third angle quantity included in the angle interval (30°, 50°] is 4.

[0146] S405, determining a target photographing device from the plurality of second photographing devices based on the fourth angle quantity.

[0147] Specifically, after the multi-cascade device determines the fourth angle quantity of the deflection angles included in the angle interval, the target angle interval is determined according to the fourth angle quantity, and then the second photographing device corresponding to the target angle interval is determined as the target photographing device.

[0148] For example, the multi-cascade device determines that the third angle quantity included in the angle interval (10°, 30°] is 1, and the third angle quantity included in the angle interval (30°, 50°] is 4, determines that the interval including the largest fourth angle quantity is the target angle interval, that is, determines that the angle interval (30°, 50°] is the target angle interval, and then determines that the photographing device F corresponding to the target angle interval is the target photographing device. By determining the angle interval including the largest fourth angle quantity as the target angle interval from the plurality of deflection angles, the target photographing device is determined, thereby improving the accuracy of determining the target photographing device.

[0149] Specifically, the multi-cascade device determines a fifth angle quantity of the deflection angles, and when the fifth angle quantity is less than a second preset angle quantity, the first image photographed by the first photographing device is reacquired until the fifth angle quantity is greater than or equal to the second preset angle quantity.

[0150] The second preset angle quantity can be a preset value for determining whether the first image needs to be reacquired.

[0151] For example, when the second preset angle number is 3, if the multi-cascade device determines that the fifth angle number of the deflection angle is 2, the first image is reacquired by the first shooting device.

[0152] In the embodiment of the present application, by reacquiring the first image when the fifth angle number of the deflection angle is less than the second preset angle number, the target shooting device is determined using multiple deflection angles, which improves the accuracy of determining the target shooting device to shoot the target person's front face image.

[0153] In the embodiment of the present application, multiple shooting devices are arranged in the target direction, and then the target shooting device is determined according to multiple deflection angles of the target person. By using different shooting devices to obtain the front face image of the target person when the deflection angle of the target person is in different angle intervals of the target direction, a more accurate front face image of the target person can be obtained.

[0154] Please refer to Figure 9 A structural diagram of a multi-cascade device is provided for the embodiment of the present application. As shown in Figure 9 The multi-cascade device 1 of the embodiment of the present application can include: a head image extraction module 11, an image determination module 12, a first shooting device determination module 13, a second shooting device determination module 14, and a front face information extraction module 15.

[0155] The head image extraction module 11 is configured to acquire multiple first images obtained by the first shooting device continuously shooting, and extract head image information of the target person from each first image based on first coordinate information of the target person in the first image.

[0156] The image determination module 12 is configured to determine a second image in which face image information of the target person does not exist in the head image information in the multiple first images.

[0157] The first shooting device determination module 13 is configured to determine the second shooting device arranged opposite to the first shooting device as the target shooting device if the image number of the second image is greater than a preset image number.

[0158] Specifically, please refer to Figure 10 A structural diagram of a feature matching module is provided for the embodiment of the present application. As shown in Figure 10 The first shooting device determination module 13 can include:

[0159] The target image acquisition unit 131 is configured to acquire multiple target images based on any second shooting device if there are multiple second shooting devices.

[0160] a deflection angle determining unit 132 for extracting target facial image information of a target person in each target image, and determining a target deflection angle of the target person relative to the second camera based on the target facial image information;

[0161] An angle interval determining unit 133 is used to determine angle intervals in which a plurality of target deflection angles are respectively located, each angle interval corresponding to one of the plurality of shooting devices;

[0162] An angle quantity determining unit 134, configured to determine a fourth angle quantity of deflection angles included in each angle interval;

[0163] The target photographing device determining unit 135 is configured to determine a target photographing device from the plurality of second photographing devices based on the fourth angle quantity.

[0164] The target shooting device determining unit 135 is specifically configured to determine the angle interval with the largest number of fourth angles in the angle interval as the target angle interval;

[0165] The second shooting device corresponding to the target angle interval is determined as the target shooting device.

[0166] a second camera determining module 14 configured to determine, if the number of images of the second image is less than or equal to a preset number of images, a deflection angle of the target person relative to the first camera based on information of each facial image, and determine a target camera in the camera set based on the plurality of deflection angles;

[0167] For details, please refer to Figure 11 , which provides a structural diagram of a feature matching module according to an embodiment of the present application. Figure 11 As shown, the second shooting device determination module 14 may include:

[0168] a variance determination unit 141, configured to determine an average angle value of the plurality of deflection angles and a variance value of the plurality of deflection angles;

[0169] an orientation determining unit 142 for determining a target orientation corresponding to a facial deflection direction of the target person based on the average angle value and the variance value;

[0170] The target photographing device determining unit 143 is configured to determine a target photographing device in the photographing device set based on the target orientation.

[0171] Optionally, the orientation determining unit 142 is specifically configured to determine the target orientation corresponding to the face deflection direction of the target person as the first orientation if the average angle value is included in the preset angle range and the variance value is less than the preset variance value;

[0172] If the average angle value is not included in the preset angle range or the variance value is greater than or equal to the preset variance value, a target orientation corresponding to the face deflection direction of the target person is determined based on the plurality of deflection angles.

[0173] Optionally, the orientation determination unit 142 is specifically configured to determine, from the plurality of deflection angles, a first deflection angle greater than a first angle threshold, and determine a first angle quantity of the first deflection angle.

[0174] If it is determined that the first angle quantity is greater than or equal to a first preset angle quantity, the target orientation corresponding to the face deflection direction of the target person is determined as a third orientation.

[0175] Or,

[0176] A second deflection angle less than a second angle threshold is determined from the plurality of deflection angles, and a second angle quantity of the second deflection angle is determined.

[0177] If it is determined that the second angle quantity is greater than or equal to the first preset angle quantity, the target orientation corresponding to the face deflection direction of the target person is determined as a fourth orientation.

[0178] The target shooting device determination unit 143 is specifically configured to, if it is determined that the target orientation is the third orientation, determine a third shooting device installed at the third orientation as the target shooting device.

[0179] If it is determined that the target orientation is the fourth orientation, a fourth shooting device installed at the fourth orientation is determined as the target shooting device.

[0180] The target shooting device determination unit 143 is specifically configured to, if it is determined that the target orientation is the first orientation, determine a first shooting device installed at the first orientation as the target shooting device.

[0181] The target shooting device determination unit 143 is specifically configured to, if there are a plurality of shooting devices in the target orientation, determine angle intervals in which the plurality of deflection angles are located, and each angle interval corresponds to one of the plurality of shooting devices.

[0182] A third angle quantity of the deflection angles included in each angle interval is determined.

[0183] Based on the third angle quantity, a target shooting device is determined from the plurality of shooting devices in the target orientation.

[0184] The target shooting device determination unit 143 is specifically configured to determine, from the angle intervals, an interval with the largest third angle quantity as a target angle interval.

[0185] The target shooting device determination unit 143 is specifically configured to determine, from the target angle interval, a shooting device corresponding to the target angle interval as the target shooting device.

[0186] The front face information extraction module 15 is configured to convert the first coordinate information into second coordinate information of a third image captured by the target photographing device, and extract a front face image of the target person from the third image based on the second coordinate information.

[0187] The first photographing device, the second photographing device, and the photographing device set are arranged in the same photographing space, the first photographing device is arranged at a first position in the photographing space, the second photographing device is arranged at a second position in the photographing space, the second position and the first position are relative positions in the photographing space, photographing devices in the photographing device set are arranged at a third position and a fourth position, respectively, the third position and the fourth position are relative positions in the photographing space, and the first position and the second position are located on two sides of a first connection line between the third position and the fourth position.

[0188] Reference Figure 12 The multi-camera cascade device further includes:

[0189] The fifth angle number determination module 16 is configured to determine a fifth angle number of the deflection angle.

[0190] The first image acquisition module 17 is configured to, when the fifth angle number is less than a second preset angle number, reacquire the first image captured by the first photographing device until the fifth angle number is greater than or equal to the second preset angle number.

[0191] In the embodiments of the present application, the target photographing device for acquiring the front face image of the target person can be determined in the case that the head image information of the target person in the captured first image does not include the face image information, and the target photographing device for acquiring the front face image of the target person can be determined according to the deflection angle determined based on the face image information of the target person in the case that the face image information of the target person exists in the first image, so that the target photographing device for acquiring the front face image of the target person can be accurately determined, the front face image of the target person is acquired through the target photographing device, the accuracy of the acquired front face image of the target person is improved, and the convenience of acquiring the front face image of the target person is improved without manual acquisition.

[0192] The embodiments of the present application also provide a computer storage medium, which can store a plurality of program instructions, the program instructions being suitable for being loaded and executed by a processor to perform the method steps of the embodiments of the present application as shown in the above Figures 1-8 The specific execution process can be referred to the specific description of the embodiments of the present application as shown in the above Figures 1-8 The specific execution process can be referred to the specific description of the embodiments of the present application as shown in the above

[0193] Please refer to Figure 13 The embodiments of the present application provide a structural schematic diagram of an electronic device. As shown in Figure 13As shown, the electronic device 1000 can include at least one processor 1001, such as a CPU, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002. The communication bus 1002 is used to realize the connection communication between the components. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a WI-FI interface). The memory 1005 can be a high-speed RAM memory, or a non-volatile memory, such as at least one disk memory. The memory 1005 can also be at least one storage device located away from the aforementioned processor 1001. As shown Figure 13 As a computer storage medium, the memory 1005 can include an operating system, a network communication module, an input / output interface module, and an information display application.

[0194] In Figure 13 As shown in the electronic device 1000, the user interface 1003 is mainly used to provide an interface for user input, and obtain user input data.

[0195] In one embodiment, the processor 1001 can be used to call the image acquisition program stored in the memory 1005, and specifically perform the following operations:

[0196] Obtain a plurality of first images obtained by continuous shooting of a first shooting device, and extract head image information of the target person from each first image based on first coordinate information of the target person in the first image;

[0197] Determine a second image in which the face image information of the target person does not exist in the head image information in the plurality of first images;

[0198] If the number of images of the second image is greater than a preset image number, determine a second shooting device arranged opposite to the first shooting device as the target shooting device;

[0199] If the number of images of the second image is less than or equal to the preset image number, determine a deflection angle of the target person relative to the first shooting device based on each face image information, and determine the target shooting device in the shooting device set based on a plurality of deflection angles;

[0200] Convert the first coordinate information into second coordinate information of a third image obtained by shooting of the target shooting device, and extract a front face image of the target person from the third image based on the second coordinate information;

[0201] The first shooting device, the second shooting device, and the shooting device set are arranged in the same shooting space, the first shooting device is arranged at a first position in the shooting space, the second shooting device is arranged at a second position in the shooting space, the second position and the first position are relative positions in the shooting space, the shooting devices in the shooting device set are arranged at a third position and a fourth position respectively, the third position and the fourth position are relative positions in the shooting space, and the first position and the second position are located on two sides of a first connection line between the third position and the fourth position.

[0202] In one embodiment, when the processor 110 determines the target shooting device in the shooting device set based on the plurality of deflection angles, the processor 110 specifically performs the following operations:

[0203] determining an average angle value of the plurality of deflection angles and a variance value of the plurality of deflection angles;

[0204] determining a target position corresponding to the face deflection direction of the target person based on the average angle value and the variance value;

[0205] determining the target shooting device in the shooting device set based on the target position.

[0206] In one embodiment, when the processor 110 determines the target position corresponding to the face deflection direction of the target person based on the average angle value and the variance value, the processor 110 specifically performs the following operations:

[0207] if the average angle value is included in a preset angle range and the variance value is less than a preset variance value, the target position corresponding to the face deflection direction of the target person is determined as the first position;

[0208] if the average angle value is not included in the preset angle range or the variance value is greater than or equal to the preset variance value, the target position corresponding to the face deflection direction of the target person is determined based on the plurality of deflection angles.

[0209] In one embodiment, when the processor 110 determines the target position corresponding to the face deflection direction of the target person based on the plurality of deflection angles, the processor 110 specifically performs the following operations:

[0210] determining a first deflection angle greater than a first angle threshold value in the plurality of deflection angles, and determining a first angle number of the first deflection angle;

[0211] if the first angle number is greater than or equal to a first preset angle number, the target position corresponding to the face deflection direction of the target person is determined as the third position;

[0212] or,

[0213] determining a second deflection angle smaller than the second angle threshold from among the plurality of deflection angles, determining a second angle number of the second deflection angle;

[0214] if the second angle number is determined to be greater than or equal to the first preset angle number, determining the target orientation corresponding to the deflection direction of the face of the target person as a fourth orientation.

[0215] In an embodiment, the processor 110, when determining the target photographing device from among the set of photographing devices based on the target orientation, specifically performs the following operations:

[0216] if the target orientation is determined to be the third orientation, determining a third photographing device installed at the third orientation as the target photographing device;

[0217] if the target orientation is determined to be the fourth orientation, determining a fourth photographing device installed at the fourth orientation as the target photographing device.

[0218] In an embodiment, the processor 110, when determining the target photographing device from among the set of photographing devices based on the target orientation, specifically performs the following operations:

[0219] if the target orientation is determined to be the first orientation, determining the first photographing device installed at the first orientation as the target photographing device.

[0220] In an embodiment, the processor 110, when determining the target photographing device from among the set of photographing devices based on the target orientation, specifically performs the following operations:

[0221] if there are a plurality of photographing devices in the target orientation, determining angle intervals in which the plurality of deflection angles are respectively located, each angle interval corresponding to one photographing device from among the plurality of photographing devices;

[0222] determining a third angle number of deflection angles included in each angle interval;

[0223] determining the target photographing device from among the plurality of photographing devices at the target orientation based on the third angle number.

[0224] In an embodiment, the processor 110, when determining the target photographing device from among the plurality of photographing devices at the target orientation based on the third angle number, specifically performs the following operations:

[0225] determining an interval in which the third angle number is the largest from among the angle intervals as a target angle interval;

[0226] determining the photographing device corresponding to the target angle interval as the target photographing device.

[0227] In an embodiment, the processor 110, when determining the second photographing device disposed opposite to the first photographing device as the target photographing device, specifically performs the following operations:

[0228] If there are multiple second photographing apparatuses, multiple target images are acquired based on any one of the second photographing apparatuses;

[0229] Target face image information of the target person in each target image is extracted, and a target deflection angle of the target person relative to the second photographing apparatus is determined based on the target face image information;

[0230] Angle intervals in which the multiple target deflection angles are located are determined, and each angle interval corresponds to one of the multiple photographing apparatuses;

[0231] A fourth angle number of deflection angles included in each angle interval is determined;

[0232] A target photographing apparatus is determined from the multiple second photographing apparatuses based on the fourth angle number.

[0233] In an embodiment, the processor 110, when determining the target photographing apparatus from the multiple second photographing apparatuses based on the fourth angle number, specifically performs the following operations:

[0234] An angle interval in which the fourth angle number is the largest is determined as a target angle interval;

[0235] A second photographing apparatus corresponding to the target angle interval is determined as the target photographing apparatus.

[0236] In an embodiment, after determining the deflection angle of the target person relative to the first photographing apparatus based on the face image information, the processor 110 further performs the following operations:

[0237] A fifth angle number of deflection angles is determined;

[0238] When the fifth angle number is less than a second preset angle number, the first image photographed by the first photographing apparatus is reacquired until the fifth angle number is greater than or equal to the second preset angle number.

[0239] In the embodiments of the present application, the target photographing apparatus that acquires the front face image of the target person can be determined in the case where the head image information of the target person in the photographed first image does not include the face image information, and the target photographing apparatus that acquires the front face image of the target person can be determined according to the deflection angle determined based on the face image information in the case where the face image information of the target person is included in the first image, so that the target photographing apparatus that acquires the front face image of the target person can be accurately determined, the front face image of the target person is acquired through the target photographing apparatus, the accuracy of the acquired front face image of the target person is improved, and the convenience of acquiring the front face image of the target person is improved.

[0240] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware. The program can be stored in a computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiment methods can be included. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), a random access memory (RAM), or the like.

[0241] The above disclosure is only the preferred embodiment of the present application, and of course cannot limit the scope of the right of the present application, so the equivalent changes made according to the claims of the present application still belong to the scope covered by the present application.

Claims

1. An image acquisition method, characterized in that, The method comprises: obtaining a plurality of first images obtained by continuous shooting of a first shooting device, and extracting head image information of a target person from each of the first images based on first coordinate information of the target person in the first images; determining a second image in which face image information of the target person does not exist in the head image information in a plurality of the first images; if the number of images of the second image is greater than a preset image number, determining a second shooting device arranged opposite to the first shooting device as a target shooting device; if the number of images of the second image is less than or equal to the preset image number, determining a deflection angle of the target person relative to the first shooting device based on each of the face image information, and determining a target shooting device in a shooting device set based on a plurality of the deflection angles; converting the first coordinate information into second coordinate information of a third image obtained by shooting of the target shooting device, and extracting a front face image of the target person from the third image based on the second coordinate information; wherein the first shooting device, the second shooting device and the shooting device set are arranged in a same shooting space, the first shooting device is arranged at a first position in the shooting space, the second shooting device is arranged at a second position in the shooting space, the second position and the first position are relative positions in the shooting space, the shooting devices in the shooting device set are arranged at a third position and a fourth position respectively, the third position and the fourth position are relative positions in the shooting space, and the first position and the second position are located on two sides of a first connection line between the third position and the fourth position.

2. The method of claim 1, wherein, The method comprises: determining an average angle value of a plurality of the deflection angles and a variance value of a plurality of the deflection angles; determining a target position corresponding to a face deflection direction of the target person based on the average angle value and the variance value; determining the target shooting device in the shooting device set based on the target position.

3. The method of claim 2, wherein, The method comprises: if the average angle value is included in a preset angle range and the variance value is less than a preset variance value, determining that a target position corresponding to the face deflection direction of the target person is a first position; if the average angle value is not included in the preset angle range or the variance value is greater than or equal to a preset variance value, determining a target position corresponding to the face deflection direction of the target person based on a plurality of the deflection angles.

4. The method of claim 3, wherein, The method comprises: determining a first deflection angle greater than a first angle threshold value in a plurality of the deflection angles, and determining a first angle number of the first deflection angle; if it is determined that the first angle number is greater than or equal to a first preset angle number, determining that a target position corresponding to the face deflection direction of the target person is a third position; or, determining a second deflection angle smaller than the second angle threshold from the plurality of deflection angles, determining a second angle number of the second deflection angle; if it is determined that the second angle number is greater than or equal to the first preset angle number, determining a target orientation corresponding to a deflection direction of the face of the target person as a fourth orientation.

5. The method of claim 4, wherein, The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if it is determined that the target orientation is the third orientation, determining a third photographing device installed at the third orientation as the target photographing device; if it is determined that the target orientation is the fourth orientation, determining a fourth photographing device installed at the fourth orientation as the target photographing device.

6. The method of claim 3, wherein, The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if it is determined that the target orientation is the first orientation, determining a first photographing device installed at the first orientation as the target photographing device.

7. The method of claim 2, wherein, The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if there are a plurality of photographing devices in the target orientation, determining a plurality of angle intervals in which the plurality of deflection angles are respectively located, each of the angle intervals corresponding to one of the plurality of photographing devices; determining a third angle number of deflection angles contained in each of the angle intervals; determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number.

8. The method of claim 7, wherein, The determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number comprises: determining an angle interval in which a third angle number of deflection angles is the largest from the angle intervals as a target angle interval; determining a photographing device corresponding to the target angle interval as the target photographing device.

9. The method of claim 1, wherein, The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if there are a plurality of photographing devices in the target orientation, determining a plurality of angle intervals in which the plurality of deflection angles are respectively located, each of the angle intervals corresponding to one of the plurality of photographing devices; determining a third angle number of deflection angles contained in each of the angle intervals; determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number. The determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number comprises: determining an angle interval in which a third angle number of deflection angles is the largest from the angle intervals as a target angle interval; 10. The method of claim 9, wherein, determining a photographing device corresponding to the target angle interval as the target photographing device. The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if there are a plurality of photographing devices in the target orientation, determining a plurality of angle intervals in which the plurality of deflection angles are respectively located, each of the angle intervals corresponding to one of the plurality of photographing devices; 11. The method of claim 1, wherein, determining a third angle number of deflection angles contained in each of the angle intervals; determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number. The determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number comprises: determining an angle interval in which a third angle number of deflection angles is the largest from the angle intervals as a target angle interval; determining a photographing device corresponding to the target angle interval as the target photographing device. The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if there are a plurality of photographing devices in the target orientation, determining a plurality of angle intervals in which the plurality of deflection angles are respectively located, each of the angle intervals corresponding to one of the plurality of photographing devices; determining a third angle number of deflection angles contained in each of the angle intervals; determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number. The determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number comprises: determining an angle interval in which a third angle number of deflection angles is the largest from the angle intervals as a target angle interval; determining a photographing device corresponding to the target angle interval as the target photographing device. The determining the target photographing device from the set of photographing devices based on the target orientation comprises: if there are a plurality of photographing devices in the target orientation, determining a plurality of angle intervals in which the plurality of deflection angles are respectively located, each of the angle intervals corresponding to one of the plurality of photographing devices; determining a third angle number of deflection angles contained in each of the angle intervals; determining the target photographing device from the plurality of photographing devices at the target orientation based on the third angle number. When the fifth angle number is less than the second preset angle number, the first image captured by the first photographing device is reacquired until the fifth angle number is greater than or equal to the second preset angle number.

12. A computer storage medium, characterized in that The computer storage medium stores a plurality of instructions, which are suitable for being loaded and executed by the processor to perform the steps of the method according to any one of claims 1-11.

13. An electronic device, comprising: Comprise: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded and executed by the processor to perform the steps of the method according to any one of claims 1-11.

14. A multi-camera cascade apparatus, characterized by, The device comprises: A head image extraction module, configured to acquire a plurality of first images captured by a first photographing device in succession, and extract head image information of a target person from each of the first images based on first coordinate information of the target person in the first images; An image determination module, configured to determine a second image in which face image information of the target person does not exist in the head image information in a plurality of the first images; A first photographing device determination module, configured to determine a second photographing device arranged opposite to the first photographing device as a target photographing device if an image number of the second image is greater than a preset image number; A second photographing device determination module, configured to determine a deflection angle of the target person relative to the first photographing device based on each of the face image information if the image number of the second image is less than or equal to the preset image number, and determine a target photographing device in a photographing device set based on a plurality of the deflection angles; A front face information extraction module, configured to convert the first coordinate information into second coordinate information of a third image captured by the target photographing device, and extract a front face image of the target person from the third image based on the second coordinate information. Wherein, the first photographing device, the second photographing device and the photographing device set are arranged in the same photographing space, the first photographing device is arranged at a first position in the photographing space, the second photographing device is arranged at a second position in the photographing space, the second position and the first position are relative positions in the photographing space, photographing devices in the photographing device set are arranged at a third position and a fourth position respectively, the third position and the fourth position are relative positions in the photographing space, and the first position and the second position are located on two sides of a first connection line between the third position and the fourth position.

Images