Boundary identification auxiliary method and device for correlation model algorithm, and electronic equipment

Abstract

The application belongs to the technical field of image processing, and relates to a boundary recognition auxiliary method and device of a correlation model algorithm and electronic equipment. The method comprises the following steps: collecting an image of a face; establishing a coordinate system; determining the position and quantity of the human image containing the face in the image; obtaining the motion state of the human image; recognizing and determining the head-shoulder position of the human; recognizing and determining the human shape position; determining the human boundary; setting an image boundary line in the image; calculating the area of the human image, and determining the optimal magnification area of the human image; establishing an activity track range model of the human image; and magnifying each human image in the image to the optimal magnification area. Through the combination of human shape recognition and head-shoulder recognition, the application can ensure that the face image maintains the optimal magnification multiple in the image, avoids the problems of face image deviation or loss, and realizes intelligent processing of human image in a single scene and multiple complex environments.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, and more specifically, relates to a boundary recognition auxiliary method, device, and electronic device based on an association model algorithm. Background Technology

[0002] When there is only one person in the captured image, the person's position in the image is not coordinated, and the size of the person in the image is unreasonable, resulting in an unclear image. When there are multiple people, the person in the middle of the image will be out of the frame, while the people next to them will be partially or completely out of the frame, which does not meet the needs of multi-person meetings or remote collaboration.

[0003] Current facial recognition algorithms, human figure recognition algorithms, head and shoulder recognition algorithms, image acquisition technology, and AI-related technology algorithms have matured. However, there is a lack of correlation between multiple algorithms, and the algorithms are relatively fixed with no training methods for improvement. A single algorithm is used permanently on a device, resulting in a poor user experience. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a boundary recognition auxiliary method, apparatus, and electronic device based on an association model algorithm.

[0005] In a first aspect, the present invention provides a boundary recognition auxiliary method for association model algorithms, comprising the following steps:

[0006] Capture images containing human faces;

[0007] Set the parameters of the image and establish a coordinate system;

[0008] The facial recognition algorithm is used to identify facial features in the image and determine the location and number of human images containing faces in the image;

[0009] Obtain the motion state of the person image;

[0010] Based on the motion state of the human figure image, combined with the position and number of the human figure images, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the human figure, and a human figure recognition algorithm is used to determine the human figure position.

[0011] The character boundary is determined based on the head and shoulder position and the human figure position;

[0012] An image boundary line is defined in the image; the image boundary line is used to constrain the range of the movement trajectory of the person image.

[0013] Obtain the coordinates of the reference point of the person image, calculate the area of ​​the person image, and determine the optimal magnified area of ​​the person image;

[0014] Based on the boundary of the person and the boundary line of the image, establish a model of the range of the person's movement trajectory;

[0015] Based on the activity trajectory range model, each of the figures in the image is magnified to the optimal magnification area.

[0016] Secondly, the present invention provides a boundary recognition auxiliary device for an association model algorithm, including a collection unit, a first setting unit, a face recognition unit, an acquisition unit, a location recognition unit, a first processing unit, a second setting unit, a second processing unit, a model building unit, and a magnification processing unit;

[0017] The acquisition unit is used to acquire images containing human faces;

[0018] The first setting unit is used to set the parameters of the image and establish a coordinate system;

[0019] The face recognition unit is used to identify facial features in the image using a face recognition algorithm, and to determine the location and number of human images containing faces in the image;

[0020] The acquisition unit is used to acquire the motion state of the person image;

[0021] The position recognition unit is used to determine the head and shoulder position of the person by using a head and shoulder recognition algorithm, and to determine the position of the person by using a human figure recognition algorithm, based on the motion state of the person image and the position and number of the person images.

[0022] The first processing unit is used to determine the boundary of the human figure based on the head and shoulder position and the human figure position;

[0023] The second setting unit is used to set an image boundary line in the image; the image boundary line is used to constrain the range of the movement trajectory of the person image;

[0024] The second processing unit is used to obtain the coordinates of the reference point of the person image, calculate the area of ​​the person image, and determine the optimal magnified area of ​​the person image;

[0025] The model building unit is used to build a model of the activity trajectory range of the person image based on the person boundary and the image boundary line;

[0026] The magnification processing unit is used to magnify each of the figures in the image to the optimal magnification area according to the activity trajectory range model.

[0027] Thirdly, the present invention provides an electronic device, comprising:

[0028] Processor and memory;

[0029] The memory is used to store computer operation instructions;

[0030] The processor is configured to execute the boundary recognition auxiliary method of the association model algorithm by invoking the computer operation instructions.

[0031] The beneficial effects of this invention are: by combining human figure recognition with head and shoulder recognition, this invention can ensure that the face image maintains the optimal magnification in the image, avoid the problem of face image misalignment or missing, and realize intelligent processing of human images in a single scene and multiple complex environments.

[0032] Based on the above technical solution, the present invention can be further improved as follows.

[0033] Furthermore, based on the motion state of the person image, combined with the position and number of the person images, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the person, and a human figure recognition algorithm is used to determine the human figure position, including:

[0034] When there is only one image of a person, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the person, and the position of the person is the same as the head and shoulder position.

[0035] When there are at least two images of a person, if the image is static, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the person, and the human figure position is the same as the head and shoulder position; if the image is moving, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the image, and a human figure recognition algorithm is used to determine the human figure position.

[0036] Furthermore, the image parameters include image resolution and image magnification.

[0037] Furthermore, the reference points of the person image include the midpoint of the person's head, the farthest points on both sides of the person, and the midpoint below the person.

[0038] Furthermore, based on the boundary of the person and the boundary line of the image, a model of the range of the person's movement trajectory is established, including:

[0039] Within the image boundary line, the human figure moves within a set range. The range of movement of the human figure in the first coordinate axis direction satisfies a first set range, and the range of movement of the human figure in the second coordinate axis direction satisfies a second set range.

[0040] Furthermore, establishing the activity trajectory range model also includes: obtaining the head, neck, shoulder, and elbow ratios;

[0041] The range of motion for neck flexion / extension, arm movement, shoulder movement, and shoulder / elbow movement are set according to the head-neck-shoulder-elbow ratio.

[0042] If the person's head is within the range of motion of the shoulders, or the range of motion of the arms and shoulders is not offset from the central axis of the person's image, or the range of motion of the neck is within a set multiple of the length of the person's head, then the person's image is not moved; otherwise, if the person's head is within the range of motion of the shoulders, or the range of motion of the arms and shoulders is offset from the central axis of the person's image, or the range of motion of the neck is within a set multiple of the length of the head, then the person's image is moved so that the person's head is within the range of motion of the shoulders, the range of motion of the arms and shoulders is within the central axis of the person's image, and the range of motion of the neck is within a set multiple of the length of the head. Attached Figure Description

[0043] Figure 1 A flowchart of the boundary recognition auxiliary method of the association model algorithm provided in Embodiment 1 of the present invention;

[0044] Figure 2 A diagram showing the human figure's position;

[0045] Figure 3 This is a schematic diagram of the boundary recognition auxiliary device for the association model algorithm provided in Embodiment 2 of the present invention;

[0046] Figure 4 This is a schematic diagram of an electronic device provided in Embodiment 3 of the present invention.

[0047] In the image: A - central figure; B - figure in close-up; C - figure in distant view; D1 - highest center point of the head; D2 - coordinates of the left hand; D3 - coordinates of the right hand; D4 - coordinates of the contact surface below the body. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0049] Example 1

[0050] As an example, see the attached document. Figure 1 As shown, to solve the above-mentioned technical problems, this embodiment provides a boundary recognition auxiliary method for association model algorithms, including the following steps:

[0051] Capture images containing human faces;

[0052] Set the image parameters and establish a coordinate system;

[0053] The face recognition algorithm is used to identify facial features in an image and determine the location and number of people in the image whose faces are present.

[0054] To capture the motion state of a person's image;

[0055] Based on the motion state of the human image, combined with the position and number of human images, the head and shoulder recognition algorithm is used to determine the position of the human head and shoulders, and the human figure recognition algorithm is used to determine the position of the human figure.

[0056] Determine the character's boundaries based on the head and shoulder positions and the human figure's position;

[0057] Set image boundary lines in the image; image boundary lines are used to constrain the range of the movement trajectory of the person in the image.

[0058] Obtain the coordinates of the reference point of the person's image, calculate the area of ​​the person's image, and determine the optimal magnified area of ​​the person's image;

[0059] Based on the boundaries of the person and the image, establish a model of the range of the person's movement trajectory.

[0060] Based on the activity trajectory range model, the images of each person in the image are magnified to the optimal magnification area.

[0061] Optionally, based on the motion state of the person image, combined with the position and number of people in the image, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the person, and a human figure recognition algorithm is used to determine the human figure position, including:

[0062] When there is only one image of a person, the head and shoulder recognition algorithm is used to determine the head and shoulder position of the person. The position of the human figure is the same as the head and shoulder position.

[0063] When there are at least two images of people, if the images are static, a head and shoulder recognition algorithm is used to determine the head and shoulder positions of the person, and the human figure position is the same as the head and shoulder position. If the images of people are in motion, a head and shoulder recognition algorithm is used to determine the head and shoulder positions of the person, and a human figure recognition algorithm is used to determine the human figure position.

[0064] In practical applications, as shown in the appendix Figure 2 As shown, the human figure recognition algorithm is used to identify and locate human figures, including:

[0065] Suppose there is a person A in the center of the image, a person B in the foreground, and a person C in the background. The positions of person A, person B, and person C are determined by their location in the image. When a person is located in the center of the image, that person is person A; when a person is located at the bottom of the image, that person is person B in the foreground; and when a person is located at the top of the image, that person is person C in the background.

[0066] The image resolution is set to 1920*1080, and the 1920 side of the image canvas is set as the X-axis and the 1080 side as the Y-axis to establish a coordinate system.

[0067] The reference points for the central figure A include four points: the highest center point of the head, D1, with coordinates (x960, y(1080*0.618)); the coordinates of the left hand, D2, are (x480, Y1); the coordinates of the right hand, D3, are (x1440, Y2); and the coordinates of the contact surface below the body, D4, are (x480, 0) - (x1440.0). Therefore, the area of ​​A is: x960*y677.44 = A. A represents the optimal maximum zoom level of the figure, which is set according to the image size. During the formation of A, the human figure's movement trajectory moves within the following coordinate system (0, 878.72) - (1920, 878.72).

[0068] Let a be the distance the character moves in the X direction, and b be the distance the character moves in the Y direction. Then the model of the humanoid's movement trajectory range is:

[0069] X / 2 ≤ a < X + (1920 - X / 2);

[0070] Y / 2 ≤ b < Y + (1080 - Y / 2);

[0071] X*Y=A is a fixed value for magnifying the figure, i.e., X=960, Y=677.44.

[0072] In practical applications, head and shoulder recognition algorithms are used to determine the head and shoulder positions in images of people, including:

[0073] The image resolution is set to 1920*1080, with the 1920 side as the X-axis and the 1080 side as the Y-axis. A coordinate system is established, and the proportions of a person's head, neck, shoulders, and elbows are obtained based on big data.

[0074] The head and shoulder model is set as follows: head length is 'a', neck range is 0-0.5a, arm range of motion is 'a', shoulder range of motion is 1.5a-2.5a, and shoulder and elbow range of motion is 1-1.5a.

[0075] Optional image parameters include image resolution and image magnification.

[0076] Optionally, the reference points for the portrait image include the midpoint of the portrait's head, the furthest points on both sides of the portrait, and the midpoint below the portrait.

[0077] Optionally, based on the boundary lines of the person and the image, a model of the range of the person's movement trajectory is established, including:

[0078] Within the image boundary, the human figure moves within a set range. The range of movement of the human figure in the first coordinate axis direction satisfies the first set range, and the range of movement of the human figure in the second coordinate axis direction satisfies the second set range.

[0079] Optionally, establishing an activity trajectory range model may also include: obtaining the head, neck, shoulder, and elbow ratios;

[0080] The range of motion for neck flexion / extension, arm movement, shoulder movement, and shoulder / elbow movement should be set according to the head-neck-shoulder-elbow ratio.

[0081] If the person's head is within the range of shoulder movement, or the range of arm and shoulder movement is not offset from the central axis of the person's image, or the range of neck movement is within a set multiple of the head length, then the person's image will not be moved. Otherwise, if the person's head is within the range of shoulder movement, or the range of arm and shoulder movement is offset from the central axis of the person's image, or the range of neck movement is within a set multiple of the head length, then the person's image will be moved so that the person's head is within the range of shoulder movement, the range of arm and shoulder movement is within the central axis of the person's image, and the range of neck movement is within a set multiple of the head length.

[0082] The head moves within a range of 1.5a and 2.5a shoulder width. The range of arm and shoulder movements does not deviate from the central axis of the image. The neck pitch range is within 0-0.5a. Within this range, no image tracking is performed, meaning the captured image area is not moved.

[0083] Boundary definition: Different resolutions are defined according to a 16:9 aspect ratio, such as 4K is 3840*2160, 2K is 2560*1440, and 1080P is 1920*1080. The image is automatically scaled on canvases of different resolutions, and the boundary is the pixel point with the maximum range of movement on the image.

[0084] This invention utilizes a head and shoulder recognition algorithm to determine the head and shoulder position of a person, and a human figure recognition algorithm to determine the human figure position. By associating the head and shoulder recognition algorithm with the human figure recognition algorithm, it can ensure that the face image maintains the optimal magnification in the image, avoid the problem of face image misalignment or missing, and realize intelligent processing of human images in a single scene and multiple complex environments.

[0085] Example 2

[0086] Based on the same principle as the method shown in Embodiment 1 of the present invention, the embodiments of the present invention also provide a boundary recognition auxiliary device for the association model algorithm, including a collection unit, a first setting unit, a face recognition unit, an acquisition unit, a location recognition unit, a first processing unit, a second setting unit, a second processing unit, a model building unit, and a magnification processing unit;

[0087] The acquisition unit is used to acquire images containing human faces;

[0088] The first setting unit is used to set the parameters of the image and establish a coordinate system;

[0089] The face recognition unit is used to identify facial features in an image using a face recognition algorithm, and to determine the location and number of people in the image whose faces are present.

[0090] The acquisition unit is used to acquire the motion state of the human image;

[0091] The position recognition unit is used to determine the head and shoulder position of a person by using a head and shoulder recognition algorithm, and to determine the position of a person by using a human figure recognition algorithm, based on the motion state of the person image, combined with the position and number of people in the image;

[0092] The first processing unit is used to determine the character boundary based on the head and shoulder position and the human figure position;

[0093] The second setting unit is used to set image boundary lines in the image; the image boundary lines are used to constrain the range of the movement trajectory of the person image.

[0094] The second processing unit is used to obtain the coordinates of the reference point of the person image, calculate the area of ​​the person image, and determine the optimal magnified area of ​​the person image.

[0095] The model building unit is used to build a model of the range of the movement trajectory of the person's image based on the person's boundary and the image boundary line;

[0096] The magnification processing unit is used to magnify each person's image in the image to the optimal magnification area based on the activity trajectory range model.

[0097] Optionally, based on the motion state of the person image, combined with the position and number of people in the image, a head and shoulder recognition algorithm is used to determine the head and shoulder position of the person, and a human figure recognition algorithm is used to determine the human figure position, including:

[0098] When there is only one image of a person, the head and shoulder recognition algorithm is used to determine the head and shoulder position of the person. The position of the human figure is the same as the head and shoulder position.

[0099] When there are at least two images of people, if the images are static, a head and shoulder recognition algorithm is used to determine the head and shoulder positions of the person, and the human figure position is the same as the head and shoulder position. If the images of people are in motion, a head and shoulder recognition algorithm is used to determine the head and shoulder positions of the person, and a human figure recognition algorithm is used to determine the human figure position.

[0100] Optional image parameters include image resolution and image magnification.

[0101] Optionally, the reference points for the portrait image include the midpoint of the portrait's head, the furthest points on both sides of the portrait, and the midpoint below the portrait.

[0102] Optionally, based on the boundary lines of the person and the image, a model of the range of the person's movement trajectory is established, including:

[0103] Within the image boundary, the human figure moves within a set range. The range of movement of the human figure in the first coordinate axis direction satisfies the first set range, and the range of movement of the human figure in the second coordinate axis direction satisfies the second set range.

[0104] Optionally, establishing an activity trajectory range model may also include: obtaining the head, neck, shoulder, and elbow ratios;

[0105] The range of motion for neck flexion / extension, arm movement, shoulder movement, and shoulder / elbow movement should be set according to the head-neck-shoulder-elbow ratio.

[0106] If the person's head is within the range of shoulder movement, or the range of arm and shoulder movement is not offset from the central axis of the person's image, or the range of neck movement is within a set multiple of the head length, then the person's image will not be moved. Otherwise, if the person's head is within the range of shoulder movement, or the range of arm and shoulder movement is offset from the central axis of the person's image, or the range of neck movement is within a set multiple of the head length, then the person's image will be moved so that the person's head is within the range of shoulder movement, the range of arm and shoulder movement is within the central axis of the person's image, and the range of neck movement is within a set multiple of the head length.

[0107] Example 3

[0108] Based on the same principles as the methods shown in the embodiments of the present invention, the embodiments of the present invention also provide an electronic device, as shown in the appendix. Figure 4 As shown, the electronic device may include, but is not limited to: a processor and a memory; the memory for storing computer programs; and the processor for executing the boundary recognition auxiliary method of the association model algorithm shown in any embodiment of the present invention by calling the computer program.

[0109] In one alternative embodiment, an electronic device is provided. Figure 4 The illustrated electronic device 40 includes a processor 410 and a memory 430. The processor 410 and the memory 430 are connected, for example, via a bus 420.

[0110] Optionally, the electronic device 40 may further include a transceiver 440, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 440 is not limited to one type, and the structure of the electronic device 40 does not constitute a limitation on the embodiments of the present invention.

[0111] Processor 410 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 410 may also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0112] Bus 420 may include a pathway for transmitting information between the aforementioned components. Bus 420 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 420 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0113] The memory 430 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0114] The memory 430 is used to store application code (computer program) for executing the present invention, and its execution is controlled by the processor 410. The processor 410 is used to execute the application code stored in the memory 430 to implement the content shown in the foregoing method embodiments.

[0115] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A boundary identification assistance method for association model algorithms, characterized in that, The method comprises the steps of: collecting an image containing a human face; setting parameters of the image to establish a coordinate system; identifying human face features in the image using a face recognition algorithm to determine the position and number of human images containing human faces in the image; obtaining the motion state of the human image; determining the head-shoulder position of a human using a head-shoulder recognition algorithm and determining the human shape position using a human shape recognition algorithm according to the motion state of the human image, the position and number of human images, including: when the number of human images is one, determining the head-shoulder position of a human using a head-shoulder recognition algorithm, and the human shape position is the same as the head-shoulder position; when the number of human images is at least two, if the human image is in a static state, determining the head-shoulder position of a human using a head-shoulder recognition algorithm, and the human shape position is the same as the head-shoulder position; if the human image is in a motion state, determining the head-shoulder position of the human image using a head-shoulder recognition algorithm, and determining the human shape position using a human shape recognition algorithm; determining the human boundary according to the head-shoulder position and the human shape position; setting an image boundary line in the image; the image boundary line is used to constrain the activity track range of the human image; obtaining the coordinates of the reference point of the human image, calculating the area of the human image, and determining the optimal magnification area of the human image; establishing an activity track range model of the human image according to the human boundary and the image boundary line; According to the activity track range model, each human image in the image is magnified to the optimal magnification area.

2. The boundary identification assistance method of the association model algorithm according to claim 1, characterized in that, The image parameters include image resolution and image magnification multiple.

3. The boundary identification assistance method of the association model algorithm according to claim 1, characterized in that, The reference points of the human image include the midpoint of the human head, the farthest points on both sides of the human body, and the midpoint below the human body.

4. The boundary identification assistance method of the association model algorithm according to claim 1, characterized in that, According to the human boundary and the image boundary line, the activity track range model of the human image is established, including: Within the image boundary line, the human image moves within a set range, the movement range of the human image in the first coordinate axis direction meets the first set range, and the movement range of the human image in the second coordinate axis direction meets the second set range.

5. The boundary identification assistance method of the association model algorithm according to claim 4, characterized in that, Establishing the activity track range model also includes obtaining the head-neck-shoulder-elbow ratio; According to the head-neck-shoulder-elbow ratio, set the neck pitch range, arm movement range, shoulder movement range, and shoulder-elbow movement range; When the head of the character is within the shoulder movement range, or the arm movement range, the shoulder movement range does not deviate from the center axis of the character image, or the neck movement range is within the set multiple range of the head length of the character, the character image is not moved; otherwise, when the head of the character is within the shoulder movement range, or the arm movement range, the shoulder movement range deviates from the center axis of the character image, or the neck movement range is within the set multiple range of the head length, the character image is moved so that the head of the character is within the shoulder movement range, the arm movement range, the shoulder movement range is in the center axis of the character image, and the neck movement range is in the set multiple range of the head length.

6. A boundary recognition assistance device for a correlation model algorithm, characterized by The method comprises the following steps: The collection unit is used for collecting an image containing a face; The first setting unit is used for setting parameters of the image and establishing a coordinate system; The face recognition unit is used for recognizing face features in the image by using a face recognition algorithm, determining positions and quantities of character images containing faces in the image; The acquisition unit is used for acquiring a motion state of the character image; The position recognition unit is used for determining a head-shoulder position of a character by using a head-shoulder recognition algorithm and determining a human form position by using a human form recognition algorithm according to the motion state of the character image, the positions and quantities of the character images, comprising: The first processing unit is used for determining a character boundary according to the head-shoulder position and the human form position; The second setting unit is used for setting an image boundary line in the image, and the image boundary line is used for restricting a movement track range of the character image; The second processing unit is used for acquiring coordinates of a reference point of the character image, calculating an area of the character image, and determining an optimal magnification area of the character image; The model establishing unit is used for establishing a movement track range model of the character image according to the character boundary and the image boundary line; The magnification processing unit is used for magnifying each character image in the image to the optimal magnification area according to the movement track range model.

7. An electronic device, comprising: The processor and the memory are included; The memory is used for storing computer operation instructions; ​ The processor is configured to execute the boundary identification auxiliary method of the correlation model algorithm by invoking the computer operation instruction. The processor is configured to execute the boundary identification auxiliary method of the correlation model algorithm by invoking the computer operation instruction.

Images