Head posture recognition and reminder method, device and storage medium

Abstract

The present invention discloses a head posture recognition and reminder method, device and storage medium, belonging to the technical field of posture recognition. The method comprises: acquiring a shooting picture of a panoramic camera in real time; recognizing the head posture of a portrait in the shooting picture, and judging whether the portrait is in a head-down state; if it is recognized that at least one portrait in the shooting picture is in a head-down state, the portrait in the head-down state is taken as a target portrait, the shooting picture of the head of the target portrait is extracted and enlarged as a target picture, and the target picture is displayed on a screen picture, so as to realize lightweight head posture calculation and recognition and improve posture recognition efficiency.

Description

Technical Field

[0001] The present invention relates to the field of gesture recognition technology, and in particular to a head gesture recognition and reminder method, device and storage medium. Background Art

[0002] In the existing recognition technology for user attention, the recognition device of the existing method for recognizing the user's head posture needs to physically adjust the angle in order to capture all the portraits, and the operation is relatively complicated. In addition, the existing technology needs to identify the user's line of sight, the distance between the user and the screen, and face recognition comparison data to recognize the user's head posture. The data volume is large, the calculation is relatively complex, and the recognized posture cannot be intuitively reminded, and the posture recognition efficiency is low. Summary of the invention

[0003] The present invention provides a head posture recognition and reminder method, device and storage medium to achieve lightweight head posture calculation and recognition and improve posture recognition efficiency.

[0004] The present invention provides a head posture recognition and reminder method, comprising:

[0005] Get the pictures taken by the panoramic camera in real time;

[0006] Recognize the head posture of the person in the captured image to determine whether the person is in a head-down state;

[0007] If it is recognized that at least one person in the captured image is in a head-down state, the person in the head-down state is taken as a target person, the captured image of the head of the target person is extracted and enlarged as a target image, and the target image is displayed on the screen.

[0008] Furthermore, the step of identifying the head posture of the person in the captured image and determining whether the person is in a head-down state is specifically as follows:

[0009] Determine the head recognition and center focus of the portrait in the captured image;

[0010] Extending a horizontal line from the center focus of the head recognition as a starting point as a horizontal standard line;

[0011] Recognize the head posture of the portrait and determine a real-time head posture recognition line of the portrait; the head posture recognition line starts from the central focus of the head recognition of the portrait and extends toward the head direction;

[0012] When the head posture recognition line deviates downward from the horizontal standard line by a preset angle or more within a preset period of time, it is determined that the portrait is in a head-down state.

[0013] Furthermore, if it is recognized that at least one person in the photographed picture is in a head-down state, the person in the head-down state is taken as a target person, the photographed picture of the head of the target person is extracted and enlarged as a target picture, and the target picture is displayed on the screen, specifically:

[0014] If the number of the target person portrait in the head-down state in the shooting picture is identified to be one, the center focus of the head of the target person portrait is identified as the focus of the shooting picture, and the shooting picture of the target person portrait is magnified by a preset multiple to obtain the target picture;

[0015] The target image is displayed at the center of the screen.

[0016] Furthermore, the method of taking the portrait in the head-down state as the target portrait, extracting and enlarging the photographed picture of the head of the target portrait as the target picture, and displaying the target picture on the screen picture also includes:

[0017] If more than one target person portrait is identified in the shooting picture and is in a head-down state, the center focus of the head of each target person portrait is used as the focus of the shooting picture, and the shooting picture of each target person portrait is magnified by a preset multiple to obtain a corresponding target picture;

[0018] Each of the target images is displayed in the center of the screen in turn at a preset time interval.

[0019] Furthermore, if it is recognized that at least one person in the photographed image is in a head-down state, the person in the head-down state is used as the target person, and the method further includes:

[0020] If it is recognized that at least one portrait in the captured image is in a head-down state, the portrait in the head-down state will be taken as the target portrait, and only the head posture of the target portrait in the captured image will be continuously recognized until all the target portraits are no longer in a head-down state, and the head posture recognition operation for all portraits in the captured image will be resumed.

[0021] Furthermore, the step of identifying the head posture of the person in the captured image and determining whether the person is in a head-down state further includes:

[0022] If it is recognized that there is no person in the photographed picture in a head-down state, the head posture of the person in the photographed picture acquired in real time continues to be recognized.

[0023] As a preferred solution, the present invention uses a panoramic camera to capture images, and there is no need to add a motor module to physically adjust the angle of the camera or move the position of the camera; the present invention only focuses on the behavior recognition of the portrait in the captured image, and calculates the angle deviation of the head posture according to the angle difference formed when the head recognition and the scientific horizontal angle of 0°, determines the behavior posture of the portrait, and after the trigger point is formed, the trigger point is proportionally enlarged in the center of the image to form an intuitive positive or reverse behavioral reminder effect; the method of the present invention does not need to be equipped with face recognition comparison, nor does it need to calculate the line of sight and the distance between the user and the screen, so as to achieve lightweight computing recognition and improve posture recognition efficiency.

[0024] Accordingly, the present invention also provides a head posture recognition and reminder device, comprising: an acquisition module, a recognition module and an amplification reminder module;

[0025] The acquisition module is used to acquire the shooting picture of the panoramic camera in real time;

[0026] The recognition module is used to recognize the head posture of the portrait in the captured image and determine whether the portrait is in a head-down state;

[0027] The zoom reminder module is used to, if it is recognized that at least one portrait in the shooting picture is in a head-down state, take the portrait in the head-down state as the target portrait, extract and zoom in the shooting picture of the head of the target portrait as the target picture, and display the target picture on the screen.

[0028] Furthermore, the recognition module includes: a head-down recognition unit, a focus recognition unit and a panoramic recognition unit;

[0029] The head-down recognition unit is used to determine the head recognition center focus of the portrait in the captured image;

[0030] Extending a horizontal line from the center focus of the head recognition as a starting point as a horizontal standard line;

[0031] Recognize the head posture of the portrait and determine a real-time head posture recognition line of the portrait; the head posture recognition line starts from the central focus of the head recognition of the portrait and extends toward the head direction;

[0032] When the head posture recognition line deviates downward from the horizontal standard line by a predetermined angle or more within a predetermined time, it is determined that the person is in a head-down state;

[0033] The focus recognition unit is used for, if it is recognized that at least one person in the photographed picture is in a head-down state, taking the person in the head-down state as a target person, and continuing to recognize only the head posture of the target person in the photographed picture until all the target persons are no longer in the head-down state, and then resuming the head posture recognition operation for all the persons in the photographed picture;

[0034] The panoramic recognition unit is used to continue to recognize the head posture of the portrait in the shooting picture acquired in real time if it is recognized that there is no portrait in the shooting picture in a head-down state.

[0035] Further, the magnification reminder module includes: a single-person magnification unit and a multi-person magnification unit;

[0036] The single person magnification unit is used for, if it is recognized that there is only one target person portrait in the shooting picture who is in a head-down state, identifying the center focus of the head of the target person portrait as the focus of the shooting picture, and magnifying the shooting picture of the target person portrait by a preset multiple to obtain the target picture;

[0037] Displaying the target image at a fixed point and in the center of the screen;

[0038] The multi-person magnification unit is used for, if it is recognized that there are more than one target portraits in the shooting picture who are in a head-down state, identifying the center focus of the head of each target portrait as the focus of the shooting picture, and magnifying the shooting picture of each target portrait by a preset multiple to obtain the corresponding target picture;

[0039] Each of the target images is displayed in the center of the screen in turn at a preset time interval.

[0040] Correspondingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, it controls the device where the computer-readable storage medium is located to execute a head posture recognition and reminder method as described in the content of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flowchart of an embodiment of the head posture recognition and reminder method provided by the present invention;

[0042] Figure 2 It is a structural schematic diagram of an embodiment of the head posture recognition and reminder device provided by the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Embodiment 1

[0045] As a head posture recognition and reminder method provided by an embodiment of the present invention, it is preferably used in a classroom focus device that assists teachers in teaching and improves students' listening quality. The panoramic camera is installed in the classroom to shoot the front of the students. There is no need to enter student information to form a database, no need to carry out face recognition comparison, and no need to divide cloud space to store presets.

[0046] Please refer to Figure 1 , a head posture recognition and reminder method provided by an embodiment of the present invention, comprising steps S101-S103:

[0047] Step S101: acquiring the shooting picture of the panoramic camera in real time;

[0048] In this embodiment, the panoramic camera does not need to add a motor module to physically adjust the angle of the camera, and does not need to move the position of the camera.

[0049] Step S102: recognizing the head posture of the person in the captured image to determine whether the person is in a head-down state;

[0050] Furthermore, the step of identifying the head posture of the person in the captured image and determining whether the person is in a head-down state is specifically as follows:

[0051] Determine the head recognition and center focus of the portrait in the captured image;

[0052] Extending a horizontal line from the center focus of the head recognition as a starting point as a horizontal standard line;

[0053] Recognize the head posture of the portrait and determine a real-time head posture recognition line of the portrait; the head posture recognition line starts from the central focus of the head recognition of the portrait and extends toward the head direction;

[0054] When the head posture recognition line deviates downward from the horizontal standard line by a preset angle or more within a preset period of time, it is determined that the portrait is in a head-down state.

[0055] In this embodiment, normal panoramic recognition is performed on the captured image, with the scientific horizontal line as the starting standard, specifically: the head of the portrait is extracted to identify the central focus, and a horizontal standard 0° line is extended as the horizontal standard line.

[0056] Using AI behavior incentive technology to continuously identify the captured image and determine the real-time head posture recognition line of the portrait;

[0057] When the head posture recognition line deviates downward from the horizontal standard line by ≥60° and lasts for 3s, it is determined that the user is in a head-down state.

[0058] Furthermore, the step of identifying the head posture of the person in the captured image and determining whether the person is in a head-down state further includes:

[0059] If it is recognized that there is no person in the photographed picture in a head-down state, the head posture of the person in the photographed picture acquired in real time continues to be recognized.

[0060] Step S103: If it is identified that at least one person in the photographed image is in a head-down state, the person in the head-down state is taken as a target person, the photographed image of the head of the target person is extracted and enlarged as a target image, and the target image is displayed on the screen.

[0061] In this embodiment, the head of the target person is used as the center focus, the screen is focused on a fixed point, the head image of the target person is enlarged by 5 times, and the enlarged head image of the target person is displayed in the center of the screen.

[0062] Furthermore, if it is recognized that at least one person in the photographed picture is in a head-down state, the person in the head-down state is taken as a target person, the photographed picture of the head of the target person is extracted and enlarged as a target picture, and the target picture is displayed on the screen, specifically:

[0063] If the number of the target person portrait in the head-down state in the shooting picture is identified to be one, the center focus of the head of the target person portrait is identified as the focus of the shooting picture, and the shooting picture of the target person portrait is magnified by a preset multiple to obtain the target picture;

[0064] The target image is displayed at the center of the screen.

[0065] Furthermore, the method of taking the portrait in the head-down state as the target portrait, extracting and enlarging the photographed picture of the head of the target portrait as the target picture, and displaying the target picture on the screen picture also includes:

[0066] If more than one target person portrait is identified in the shooting picture and is in a head-down state, the center focus of the head of each target person portrait is used as the focus of the shooting picture, and the shooting picture of each target person portrait is magnified by a preset multiple to obtain a corresponding target picture;

[0067] Each of the target images is displayed in the center of the screen in turn at a preset time interval.

[0068] Furthermore, if it is recognized that at least one person in the photographed image is in a head-down state, the person in the head-down state is used as the target person, and the method further includes:

[0069] If it is recognized that at least one portrait in the captured image is in a head-down state, the portrait in the head-down state will be taken as the target portrait, and only the head posture of the target portrait in the captured image will be continuously recognized until all the target portraits are no longer in a head-down state, and the head posture recognition operation for all portraits in the captured image will be resumed.

[0070] In this embodiment, the user's head-down state is divided into two situations:

[0071] One is to automatically focus on the head of a person who is looking down at books and materials, and enlarge the head image of the person and display it on the screen, which is a positive praise and motivation;

[0072] The other is to automatically focus on the head of the person who is dozing off or distracted by the head, and enlarge the image of the head and display it on the screen, which serves as a reverse reminder and motivation.

[0073] Implementing the embodiments of the present invention has the following effects:

[0074] The present invention uses a panoramic camera to shoot pictures, and there is no need to add a motor module to physically adjust the angle of the camera or move the position of the camera; the present invention only focuses on the behavior recognition of the portrait in the shot picture, calculates the angle deviation of the head posture according to the angle difference formed when the head recognition and the scientific horizontal angle of 0°, determines the behavior posture of the portrait, and after the trigger point is formed, the trigger point is proportionally enlarged in the center of the picture to form an intuitive positive or reverse behavior reminder; the method of the present invention does not need to be equipped with face recognition comparison, nor does it need to calculate the line of sight and the distance between the user and the screen, so as to achieve lightweight calculation and recognition and improve the efficiency of posture recognition.

[0075] Embodiment 2

[0076] Please refer to Figure 2 , a head posture recognition and reminder device provided by an embodiment of the present invention includes: an acquisition module 201, a recognition module 202 and an amplification reminder module 203;

[0077] The acquisition module is used to acquire the shooting picture of the panoramic camera in real time;

[0078] The recognition module is used to recognize the head posture of the portrait in the captured image and determine whether the portrait is in a head-down state;

[0079] The zoom reminder module is used to, if it is recognized that at least one portrait in the shooting picture is in a head-down state, take the portrait in the head-down state as the target portrait, extract and zoom in the shooting picture of the head of the target portrait as the target picture, and display the target picture on the screen.

[0080] Furthermore, the recognition module includes: a head-down recognition unit, a focus recognition unit and a panoramic recognition unit;

[0081] The head-down recognition unit is used to determine the head recognition center focus of the portrait in the captured image;

[0082] Extending a horizontal line from the center focus of the head recognition as a starting point as a horizontal standard line;

[0083] Recognize the head posture of the portrait and determine a real-time head posture recognition line of the portrait; the head posture recognition line starts from the central focus of the head recognition of the portrait and extends toward the head direction;

[0084] When the head posture recognition line deviates downward from the horizontal standard line by a predetermined angle or more within a predetermined time, it is determined that the person is in a head-down state;

[0085] The focus recognition unit is used for, if it is recognized that at least one person in the photographed picture is in a head-down state, taking the person in the head-down state as a target person, and continuing to recognize only the head posture of the target person in the photographed picture until all the target persons are no longer in the head-down state, and then resuming the head posture recognition operation for all the persons in the photographed picture;

[0086] The panoramic recognition unit is used to continue to recognize the head posture of the portrait in the shooting picture acquired in real time if it is recognized that there is no portrait in the shooting picture in a head-down state.

[0087] Further, the magnification reminder module includes: a single-person magnification unit and a multi-person magnification unit;

[0088] The single person magnification unit is used for, if it is recognized that there is only one target person portrait in the shooting picture who is in a head-down state, identifying the center focus of the head of the target person portrait as the focus of the shooting picture, and magnifying the shooting picture of the target person portrait by a preset multiple to obtain the target picture;

[0089] Displaying the target image at a fixed point and in the center of the screen;

[0090] The multi-person magnification unit is used for, if it is recognized that there are more than one target portraits in the shooting picture who are in a head-down state, identifying the center focus of the head of each target portrait as the focus of the shooting picture, and magnifying the shooting picture of each target portrait by a preset multiple to obtain the corresponding target picture;

[0091] Each of the target images is displayed in the center of the screen in turn at a preset time interval.

[0092] The above-mentioned head posture recognition and reminder device can implement the head posture recognition and reminder method of the above-mentioned method embodiment. The optional items in the above-mentioned method embodiment are also applicable to this embodiment and will not be described in detail here. The rest of the contents of the embodiment of this application can refer to the contents of the above-mentioned method embodiment, and will not be repeated in this embodiment.

[0093] Embodiment 3

[0094] Correspondingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the head posture recognition and reminder method described in any one of the above embodiments.

[0095] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.

[0096] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0097] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0098] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile terminal, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (SecureDi gita l, SD) card, a flash card (F l ash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0099] Wherein, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0100] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A head posture recognition and reminder method, characterized in that: include: Get the pictures taken by the panoramic camera in real time; Recognize the head posture of the person in the captured image to determine whether the person is in a head-down state, specifically: Determine the head recognition and center focus of the portrait in the captured image; Extending a horizontal line from the center focus of the head recognition as a starting point as a horizontal standard line; Recognize the head posture of the portrait and determine the real-time head posture recognition line of the portrait; The head posture recognition line starts from the central focus of the head recognition of the portrait and extends towards the head; When the head posture recognition line deviates downward from the horizontal standard line by a predetermined angle or more within a predetermined time, it is determined that the person is in a head-down state; If it is recognized that at least one person in the captured image is in a head-down state, the person in the head-down state is taken as a target person, the captured image of the head of the target person is extracted and enlarged as a target image, and the target image is displayed on the screen.

2. A head posture recognition and reminder method as claimed in claim 1, characterized in that: If it is recognized that at least one person in the photographed picture is in a head-down state, the person in the head-down state is taken as a target person, a photographed picture of the head of the target person is extracted and enlarged as a target picture, and the target picture is displayed on the screen, specifically: If the number of the target person portrait in the head-down state in the shooting picture is identified to be one, the center focus of the head of the target person portrait is identified as the focus of the shooting picture, and the shooting picture of the target person portrait is magnified by a preset multiple to obtain the target picture; The target image is displayed at the center of the screen.

3. A head posture recognition and reminder method as claimed in claim 2, characterized in that: The method of taking the portrait in the head-down state as the target portrait, extracting and enlarging the photographed picture of the head of the target portrait as the target picture, and displaying the target picture on the screen picture also includes: If more than one target person portrait is identified in the shooting picture and is in a head-down state, the center focus of the head of each target person portrait is used as the focus of the shooting picture, and the shooting picture of each target person portrait is magnified by a preset multiple to obtain a corresponding target picture; Each of the target images is displayed in the center of the screen in turn at a preset time interval.

4. A head posture recognition and reminder method as claimed in claim 1, characterized in that: If it is recognized that at least one person in the photographed image is in a head-down state, the person in the head-down state is used as a target person, and the method further includes: If it is recognized that at least one portrait in the captured image is in a head-down state, the portrait in the head-down state will be taken as the target portrait, and only the head posture of the target portrait in the captured image will be continuously recognized until all the target portraits are no longer in a head-down state, and the head posture recognition operation for all portraits in the captured image will be resumed.

5. A head posture recognition and reminder method as claimed in claim 1, characterized in that: The step of recognizing the head posture of the person in the photographed image and determining whether the person is in a head-down state further includes: If it is recognized that there is no person in the photographed picture in a head-down state, the head posture of the person in the photographed picture acquired in real time continues to be recognized.

6. A head posture recognition and reminder device, characterized in that: include: Acquisition module, recognition module and magnification reminder module; The acquisition module is used to acquire the shooting picture of the panoramic camera in real time; The recognition module is used to recognize the head posture of the portrait in the captured image and determine whether the portrait is in a head-down state; The recognition module includes: a head-down recognition unit, a focus recognition unit and a panoramic recognition unit; The head-down recognition unit is used to determine the head recognition center focus of the portrait in the captured image; Extending a horizontal line from the center focus of the head recognition as a starting point as a horizontal standard line; Recognize the head posture of the portrait and determine a real-time head posture recognition line of the portrait; the head posture recognition line starts from the central focus of the head recognition of the portrait and extends toward the head direction; When the head posture recognition line deviates downward from the horizontal standard line by a predetermined angle or more within a predetermined time, it is determined that the person is in a head-down state; The focus recognition unit is used for, if it is recognized that at least one person in the photographed picture is in a head-down state, taking the person in the head-down state as a target person, and continuing to recognize only the head posture of the target person in the photographed picture until all the target persons are no longer in the head-down state, and then resuming the head posture recognition operation for all the persons in the photographed picture; The panoramic recognition unit is used to continue to recognize the head posture of the portrait in the real-time captured shooting picture if it is recognized that there is no portrait in the shooting picture in a head-down state; The enlargement reminder module is used to, if it is recognized that at least one portrait in the shooting picture is in a head-down state, take the portrait in the head-down state as the target portrait, extract and enlarge the shooting picture of the head of the target portrait as the target picture, and display the target picture on the screen.

7. A head posture recognition and reminder device as claimed in claim 6, characterized in that: The magnification reminder module includes: a single-person magnification unit and a multi-person magnification unit; The single person magnification unit is used for, if it is recognized that there is only one target person portrait in the shooting picture who is in a head-down state, identifying the center focus of the head of the target person portrait as the focus of the shooting picture, and magnifying the shooting picture of the target person portrait by a preset multiple to obtain the target picture; Displaying the target image at a fixed point and in the center of the screen; The multi-person magnification unit is used for, if it is recognized that there are more than one target portraits in the shooting picture who are in a head-down state, identifying the center focus of the head of each target portrait as the focus of the shooting picture, and magnifying the shooting picture of each target portrait by a preset multiple to obtain the corresponding target picture; Each of the target images is displayed in the center of the screen in turn at a preset time interval.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program; wherein, when the computer program is run, it controls the device where the computer-readable storage medium is located to execute a head posture recognition and reminder method as described in any one of claims 1 to 5.

Images