A human shape movement monitoring method, system, terminal and storage medium

Abstract

The application relates to a human shape movement monitoring method, a system, a terminal and a storage medium, and belongs to the technical field of monitoring, wherein the method is based on a processing terminal in a monitoring device, the monitoring device further comprises a camera and a driving piece for adjusting the angle of the camera, the method comprises the following steps: acquiring real-time monitoring images; performing image recognition processing on the real-time monitoring images; confirming a target following object and entering a lens following mode after the recognition result meets preset mode switching conditions; and performing the following processing in the lens following mode: identifying the position change trend of the target following object; and generating a control instruction for the driving piece based on the position change trend to adjust the rotation angle of the camera. The application can improve the problem that the monitoring effect on specific objects is poor.

Description

Technical Field

[0001] This invention relates to the field of monitoring technology, and in particular to a method, system, terminal, and storage medium for monitoring human movement. Background Technology

[0002] With societal development and continuous advancements in science and technology, people's security awareness has also significantly increased, leading to the gradual growth of the surveillance equipment industry. As the contemporary social network environment continues to improve, user demand for surveillance equipment is increasing, ushering in a new era of digitalization and increasing convenience for surveillance equipment.

[0003] The typical monitoring method of surveillance equipment is to monitor the application scene by capturing images of the current application scene from a fixed location, without adjusting the position of the equipment itself.

[0004] The inventors have found that the above-mentioned prior art has at least the following problems: Since most people in the current application scenario are in a moving state, the monitoring angle cannot be adjusted when monitoring people in a moving state in the current application scenario using general monitoring methods. Therefore, the monitoring effect on specific objects in real-time monitoring images is not good. Summary of the Invention

[0005] To improve the problem of poor monitoring effectiveness for specific objects, this application provides a human movement monitoring method, system, terminal, and storage medium.

[0006] Firstly, this application provides a method for detecting human movement, employing the following technical solution:

[0007] A method for detecting human movement, the method being based on a processing terminal in a monitoring device, the monitoring device further including a camera and a driving component for adjusting the angle of the camera, the method comprising:

[0008] Acquire real-time monitoring images;

[0009] The real-time monitoring images are subjected to image recognition processing;

[0010] After the recognition result meets the preset mode switching conditions, the target object is confirmed and the camera follows mode is entered.

[0011] Within the camera follow mode, the following processing is performed:

[0012] Identify the positional change trend of the target object being followed;

[0013] Based on the position change trend, control commands are generated for the drive components to adjust the rotation angle of the camera.

[0014] By adopting the above technical solution, during daily monitoring, the processing terminal can automatically acquire real-time monitoring images through the camera and perform image recognition processing on these images. After the recognition results of the real-time monitoring images meet preset mode switching conditions, the processing terminal can further confirm the target object and enter lens-following mode. In lens-following mode, the processing terminal identifies the positional change trend of the target object and, based on this trend, generates control commands for the drive components. By controlling the drive components, the camera is driven to rotate as the target object moves, thus improving the monitoring effect on specific objects.

[0015] Optionally, after the recognition result meets the preset mode switching conditions, determining the target object to be followed and entering the camera follow mode specifically includes:

[0016] When the characteristics of a first person in the real-time monitoring image match the preset characteristics of key personnel, the first person is identified as the target to be followed, and the camera follows the image.

[0017] By adopting the above technical solution, the processing terminal can identify the characteristics of people appearing in real-time monitoring images. When the characteristics of a first person are identified and match any pre-stored key personnel characteristics, it indicates that the identification result of the real-time monitoring image meets the mode switching conditions. At this time, the processing terminal can confirm the first person as the target to be followed and enter the camera follow mode. By comparing the characteristics of the first person with the characteristics of key personnel, the processing terminal can automatically identify the personnel that need to be monitored and control the camera to follow and monitor them.

[0018] Optionally, after the recognition result meets the preset mode switching conditions, determining the target object to be followed and entering the camera follow mode specifically includes:

[0019] When the behavior characteristics of a second person in the real-time monitoring video are identified as matching the preset abnormal behavior characteristics, the second person is identified as the target to be followed, and the camera follows the video.

[0020] By adopting the above technical solution, the processing terminal can identify the behavioral characteristics of people appearing in real-time monitoring images. When the behavioral characteristics of a second person match any pre-stored key abnormal behavioral characteristics, it indicates that the identification result of the real-time monitoring image meets the mode switching conditions. At this time, the processing terminal can confirm the second person as the target to be followed and enter the camera follow mode. By comparing behavioral characteristics and abnormal behavioral characteristics, the processing terminal can automatically identify the personnel that need to be monitored and control the camera to follow and monitor them.

[0021] Optionally, after the recognition result meets the preset mode switching conditions, determining the target object to be followed and entering the camera follow mode specifically includes:

[0022] If the current time is within the preset control period, and a person's image is detected in the real-time monitoring image, the third person corresponding to the person's image is identified as the target to be followed, and the camera follows the image.

[0023] By adopting the above technical solution, the processing terminal has a preset control time period. When it is detected that the current time has entered the range of the preset control time period, if the processing terminal detects a person's image in the real-time monitoring video, it will identify the third person corresponding to the person's image as the target to be followed and enter the camera follow mode.

[0024] Furthermore, after identifying the third person corresponding to the person graphic as the target follower, the process also includes:

[0025] Generate prompts and send them to the preset voice prompt components;

[0026] If it is detected that a person image corresponding to the third person is always present in the real-time monitoring image within the preset waiting time, then the facial data of the third person corresponding to the person image is identified.

[0027] The facial data is added to a pre-set database of intruders for future reference.

[0028] By adopting the above technical solution, the processing terminal can start timing after confirming the third person as the target to be followed. If a person image corresponding to the third person is always present in the real-time monitoring image during the above waiting time, the processing terminal will identify the face data of the third person corresponding to the person image and add the identified face data of the third person to a preset intruder database. When a query command for an intruder is received, the processing terminal can retrieve the stored face data from the intruder database and provide feedback.

[0029] Optionally, the real-time monitoring image is pre-set with a virtual decision box, which includes decision boundaries corresponding to different directions;

[0030] The identification of the positional change trend of the target object being followed specifically includes:

[0031] When the distance between the target object being followed and the target determination boundary is less than a preset distance threshold, it is determined that the target object being followed has a positional change trend toward the direction corresponding to the target determination boundary.

[0032] Optionally, after identifying the positional change trend of the target-following object, the method further includes:

[0033] Identify the real-time movement speed of the target object being followed;

[0034] The generation of control commands for the drive component based on the position change trend specifically includes:

[0035] Based on the position change trend and real-time movement speed, control commands for the drive components are generated.

[0036] Secondly, this application provides a human figure detection system, which adopts the following technical solution:

[0037] A monitoring device includes a processing terminal, a camera, and a driving component for adjusting the angle of the camera. The processing terminal includes:

[0038] The surveillance image acquisition module is used to acquire real-time surveillance images;

[0039] The image recognition module is used to perform image recognition processing on the real-time monitoring images;

[0040] The target following object confirmation module is used to confirm the target following object after the recognition result meets the preset mode switching conditions.

[0041] The working mode switching module is used to enable the processing terminal to enter the lens following mode after the recognition result meets the preset mode switching conditions.

[0042] The position change trend recognition module is used to identify the position change trend of the target following object;

[0043] The device control module is used to generate control commands for the drive components based on the position change trend, so as to adjust the rotation angle of the camera.

[0044] Thirdly, this application provides a smart terminal, which adopts the following technical solution:

[0045] A smart terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in the first aspect.

[0046] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0047] A computer-readable storage medium includes a computer program stored thereon that can be loaded by a processor and execute the methods described in the first aspect.

[0048] In summary, this application includes at least one of the following beneficial technical effects:

[0049] 1. During routine monitoring, the processing terminal can automatically acquire real-time monitoring images through the camera and perform image recognition processing on these images. Once the recognition results of the real-time monitoring images meet preset mode switching conditions, the processing terminal can further confirm the target object and enter lens-following mode. In lens-following mode, the processing terminal identifies the positional change trend of the target object and, based on this trend, generates control commands for the drive mechanism. By controlling the drive mechanism, the camera is driven to rotate as the target object moves, thus improving the monitoring effect on specific objects.

[0050] 2. The processing terminal can identify the characteristics of people appearing in real-time monitoring images. When the characteristics of a first person match any pre-stored key personnel characteristic, it indicates that the identification result of the real-time monitoring image meets the mode switching conditions. At this time, the processing terminal can confirm the first person as the target to be followed and enter the camera follow mode. By comparing the characteristics of the first person with the characteristics of key personnel, the processing terminal can automatically identify the personnel that need to be monitored and control the camera to follow and monitor them. Attached Figure Description

[0051] Figure 1 This is a structural block diagram illustrating a monitoring device in the embodiments of this application;

[0052] Figure 2 This is a flowchart illustrating a human movement monitoring method in the embodiments of this application;

[0053] Figure 3 This is a structural block diagram illustrating the processing terminal in the embodiments of this application.

[0054] Explanation of reference numerals in the attached diagram: 11. Monitoring image acquisition module; 12. Image recognition module; 13. Target tracking object confirmation module; 14. Working mode switching module; 15. Position change trend recognition module; 16. Equipment control module. Detailed Implementation

[0055] The following is in conjunction with the appendix Figure 1-3 This application will be described in further detail.

[0056] This application discloses a method for detecting human movement, referring to... Figure 1 This method is based on a processing terminal in a monitoring device and can be applied to daily monitoring in various areas. The monitoring device also includes a camera for acquiring images and a drive mechanism for adjusting the camera angle. The drive mechanism can be an electromechanical device such as a motor.

[0057] The following will describe the specific implementation methods. Figure 2 The processing flow shown is explained in detail below:

[0058] S101: Acquire real-time monitoring images.

[0059] In practice, the processing terminal in the monitoring equipment can automatically acquire real-time monitoring images through the camera.

[0060] S102: Perform image recognition processing on real-time monitoring images.

[0061] In practice, the processing terminal performs image recognition processing on the acquired real-time monitoring images to identify the human figures within them.

[0062] S103: After the recognition result meets the preset mode switching conditions, confirm the target object to be followed and enter the lens follow mode.

[0063] In practice, the processing terminal can determine whether the recognition result of the real-time monitoring image meets the preset mode switching conditions based on the recognition result. After determining that the recognition result meets the mode switching conditions, the processing terminal can further identify the target object to be followed from the real-time monitoring image and enter the lens following mode.

[0064] Optionally, in another embodiment, the above-mentioned mode switching condition may be: identifying a person's characteristics that match any key personnel characteristic in a preset key characteristic database. The key characteristic database records key personnel characteristics that require special attention, such as specific faces, specific clothing that reflects a person's identity, etc. In this case, S103 may specifically include the following:

[0065] When a first person matching the preset key personnel characteristics is identified in the real-time monitoring video, the first person is confirmed as the target to be followed, and the camera follows the video.

[0066] In practice, the processing terminal can identify all personnel images in the real-time monitoring video, and then identify the characteristics of each personnel image. For example, it can identify the facial features, clothing features, etc. of the personnel corresponding to the personnel images, and match the identified personnel features with the key personnel features in the preset key feature library one by one. When the personnel features of the first person are identified and match any key personnel feature in the key feature library, it means that the identification result of the real-time monitoring video meets the mode switching conditions. At this time, the processing terminal can confirm the first person as the target to be followed and enter the camera follow mode.

[0067] Optionally, in another embodiment, the above-mentioned mode switching condition can also be: identifying that a person's behavioral characteristics match any abnormal behavioral characteristic in a preset abnormal behavior characteristic database. The abnormal behavior characteristic database records abnormal behavioral characteristics that require special attention, such as instances of interpersonal conflict. In this case, S103 may specifically include the following:

[0068] When a second person's behavioral characteristics are detected in the real-time monitoring video and match the preset abnormal behavioral characteristics, the second person is identified as the target to be followed, and the camera follows the video.

[0069] In practice, the processing terminal can identify all the human figures in the real-time monitoring video. After identifying the behavioral characteristics of each human figure, the processing terminal can compare the identified behavioral characteristics with each abnormal behavioral characteristic in the preset abnormal behavior feature library one by one. When the behavioral characteristics of a second human figure are identified and match any abnormal behavioral characteristic in the abnormal behavior feature library, it means that the identification result of the real-time monitoring video meets the mode switching conditions. At this time, the processing terminal can confirm the second human figure as the target to be followed and enter the camera follow mode.

[0070] Optionally, in another embodiment, the above-mentioned mode switching condition can also be: the current time is within a preset control time period, and a person's image is detected in the real-time monitoring video. The control time can be preset, or it can be set by the processing terminal upon receiving a control time setting signal. In this case, S103 may specifically include the following:

[0071] If the current time is within the preset control period and a person's image appears in the real-time monitoring video, the third person corresponding to the person's image will be identified as the target to be followed, and the camera will enter the follow mode.

[0072] In practice, the processing terminal can enter the control mode when it detects that the current time has entered the control period range. In the control mode, if the processing terminal detects a person's image in the real-time monitoring video, it will identify the third person corresponding to the person's image as the target to be followed and enter the camera follow mode until the current time exceeds the control period range, at which point it will exit the control mode.

[0073] Furthermore, in another embodiment, the monitoring device further includes a voice prompt component, which may be a speaker. After confirming the third person corresponding to the person graphic as the target to be followed, the above method may further include the following:

[0074] Generate prompts and send them to the preset voice prompt component.

[0075] In practice, the processing terminal can generate prompts and send them to the voice prompt component, so that the voice prompt component can emit a sound, which can be a preset prompt statement, thereby reminding a third party to leave the area.

[0076] If a person image corresponding to a third person is consistently detected in the real-time monitoring video within a preset waiting time, then the facial data of the third person corresponding to the person image is identified.

[0077] The waiting time can be a preset duration such as 5 seconds or 10 seconds, to allow people who receive the voice prompt to leave in a timely manner.

[0078] In practice, the processing terminal can start timing after confirming the third person as the target to be followed. If a person image corresponding to the third person is always present in the real-time monitoring image during the aforementioned waiting period, the processing terminal will identify the facial data of the third person corresponding to the person image.

[0079] Add facial data to a pre-existing database of intruders for future reference.

[0080] In practice, the processing terminal can add the facial data of the identified third party to a pre-set database of intruders. When a query command for an intruder is received, the processing terminal can retrieve the stored facial data from the database and provide feedback.

[0081] During the duration of the aforementioned camera-following mode, the processing terminal may perform the following processing:

[0082] S104: Identify the positional change trend of the target following the object.

[0083] In practice, the processing terminal can identify the positional change trend of the target following the object.

[0084] S105: Based on the position change trend, generate control commands for the drive components to adjust the rotation angle of the camera.

[0085] In practice, the processing terminal can generate control commands for the drive unit based on the identified position change trend, and control the drive unit to operate based on the control commands, so that the drive unit drives the camera to rotate, thereby enabling the camera to rotate with the position change trend to maintain monitoring of the target object.

[0086] Optionally, in another embodiment, a virtual decision box may be preset in the real-time monitoring image. This virtual decision box can be rectangular and includes decision boundaries corresponding to different directions. In this case, the above-mentioned S104 may specifically include the following:

[0087] When the distance between the target object being followed and the target determination boundary is less than a preset distance threshold, it is determined that the target object being followed has a positional change trend toward the direction corresponding to the target determination boundary.

[0088] In implementation, the processing terminal can identify the center of the human figure of the target object in real-time monitoring images. For ease of identification, the center of the portion corresponding to the head in the human figure graphic can be identified as the center of the human figure. The processing terminal then calculates the distance between the center of the human figure and each boundary of the virtual decision box in real time and compares it with preset distance thresholds. When the distance between the center of the human figure and any decision boundary is less than the aforementioned distance threshold, the processing terminal identifies that decision boundary as the target decision boundary and determines that the target object is exhibiting a positional change trend towards the direction corresponding to the target decision boundary.

[0089] Optionally, in another embodiment, after S104, the above method may further include the following:

[0090] Identify the real-time movement speed of the target object.

[0091] In practice, the processing terminal can identify the real-time movement speed of the target object.

[0092] In this case, S105 specifically includes:

[0093] Based on the position change trend and real-time movement speed, control commands for the drive components are generated.

[0094] In implementation, the processing terminal can generate control commands for the drive component based on the positional change trend and real-time movement speed of the target object. These control commands can record the rotation direction and speed of the drive component. Subsequently, the processing terminal can control the drive component based on these control commands, causing it to move the camera to rotate along with the target object.

[0095] The implementation principle of this application embodiment is as follows: During daily monitoring, the processing terminal can automatically acquire real-time monitoring images through the camera and perform image recognition processing on the real-time monitoring images. After the recognition result of the real-time monitoring image meets the preset mode switching conditions, the processing terminal can further confirm the target object and enter the lens follow mode. After entering the lens follow mode, the processing terminal will identify the position change trend of the target object and, based on the position change trend, generate control commands for the drive components. By controlling the drive components, the camera is driven to rotate as the target object moves, which helps to improve the monitoring effect of specific objects.

[0096] Based on the above method, this application also discloses a monitoring device, which includes a processing terminal, a camera, and a driving component for adjusting the camera angle. (See also...) Figure 3 The processing terminal includes:

[0097] The monitoring image acquisition module 11 is used to acquire real-time monitoring images.

[0098] Image recognition module 12 is used to perform image recognition processing on real-time monitoring images.

[0099] The target following object confirmation module 13 is used to confirm the target following object after the recognition result meets the preset mode switching conditions.

[0100] The working mode switching module 14 is used to enable the processing terminal to enter the lens following mode after the recognition result meets the preset mode switching conditions.

[0101] The position change trend recognition module 15 is used to identify the position change trend of the target following object.

[0102] The device control module 16 is used to generate control commands for the drive components based on the position change trend, so as to adjust the rotation angle of the camera.

[0103] Optionally, the target tracking object confirmation module is specifically used to confirm the first person as the target tracking object when the characteristics of the first person in the real-time monitoring image match the preset key personnel characteristics.

[0104] The working mode switching module is specifically used to enable the processing terminal to enter the lens following mode after the target following object confirmation module has confirmed the target following object.

[0105] Optionally, the target tracking object confirmation module is specifically used to identify the second person as the target tracking object when the behavioral characteristics of the second person in the real-time monitoring video match the preset abnormal behavioral characteristics.

[0106] The working mode switching module is specifically used to enable the processing terminal to enter the lens following mode after the target following object confirmation module has confirmed the target following object.

[0107] Optionally, the target following object confirmation module is specifically used to confirm the third person corresponding to the person's image as the target following object if the current time is within a preset control period and a person's image is detected in the real-time monitoring video.

[0108] The working mode switching module is specifically used to enable the processing terminal to enter the lens following mode after the target following object confirmation module has confirmed the target following object.

[0109] Optionally, the monitoring equipment may also include:

[0110] The prompt instruction generation module is used to generate prompt instructions and send them to a preset voice prompt component after confirming the third person corresponding to the person graphic as the target follower.

[0111] The image recognition module is also used to identify the face data of the third person if a person image corresponding to a third person is always present in the real-time monitoring image within a preset waiting time.

[0112] The data storage module is used to add facial data to a pre-set database of intruders for future reference.

[0113] Optionally, the monitoring equipment may also include:

[0114] The position change trend recognition module is specifically used to determine that when the distance between the target following object and the target determination boundary is less than a preset distance threshold, the target following object has a position change trend that tends to the direction corresponding to the target determination boundary.

[0115] Optionally, the monitoring equipment may also include:

[0116] The speed calculation module is used to identify the real-time movement speed of the target object being followed.

[0117] The equipment control module is specifically used to generate control commands for the drive components based on the position change trend and real-time movement speed.

[0118] This application also discloses a smart terminal, which includes a memory and a processor. The memory stores a computer program that can be loaded by the processor and executed as described above for humanoid movement monitoring.

[0119] This application also discloses a computer-readable storage medium that stores a computer program that can be loaded by a processor and executed as described above for humanoid movement monitoring. The computer-readable storage medium includes, for example, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0120] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit the scope of protection of the application. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

Claims

1. A method for detecting human movement, characterized in that, The method is based on a processing terminal in a monitoring device, which further includes a camera and a drive mechanism for adjusting the angle of the camera. The method includes: Acquire real-time monitoring images; The real-time monitoring images are subjected to image recognition processing; After the recognition result meets the preset mode switching conditions, the target object is confirmed and the camera follows the mode. Within the camera follow mode, the following processing is performed: The center of the portion corresponding to the head of the person in the graphic representation of the target being followed is identified as the center of the human figure. The real-time monitoring image is pre-set with a virtual decision box, which includes decision boundaries corresponding to different directions; Identifying the positional change trend of the target object being followed specifically includes: Calculate the distance between the center of the human figure and each of the decision boundaries of the virtual decision box; When the distance between the center of the human figure and the target determination boundary is less than a preset distance threshold, it is determined that the target following object has a positional change trend toward the direction corresponding to the target determination boundary; Based on the position change trend, control commands are generated for the drive components to adjust the rotation angle of the camera.

2. The human movement monitoring method according to claim 1, characterized in that, After the recognition result meets the preset mode switching conditions, the target object to be followed is determined and the camera-following mode is entered, which specifically includes: When the characteristics of a first person in the real-time monitoring image match the preset characteristics of key personnel, the first person is identified as the target to be followed, and the camera follows the image.

3. The human movement monitoring method according to claim 1, characterized in that, After the recognition result meets the preset mode switching conditions, the target object to be followed is determined and the camera-following mode is entered, which specifically includes: When the behavior characteristics of a second person in the real-time monitoring video are identified as matching the preset abnormal behavior characteristics, the second person is identified as the target to be followed, and the camera follows the video.

4. The human movement monitoring method according to claim 1, characterized in that, After the recognition result meets the preset mode switching conditions, the target object to be followed is determined and the camera-following mode is entered, which specifically includes: If the current time is within the preset control period, and a person's image is detected in the real-time monitoring image, the third person corresponding to the person's image is identified as the target to be followed, and the camera follows the image.

5. The human movement monitoring method according to claim 4, characterized in that, After identifying the third person corresponding to the person graphic as the target follower, the method further includes: Generate prompts and send them to the preset voice prompt components; If it is detected that a person image corresponding to the third person is always present in the real-time monitoring image within the preset waiting time, then the facial data of the third person corresponding to the person image is identified. The facial data is added to a pre-set database of intruders for future reference.

6. The human movement monitoring method according to claim 1, characterized in that, After identifying the positional change trend of the target-following object, the method further includes: Identify the real-time movement speed of the target object being followed; The generation of control commands for the drive component based on the position change trend specifically includes: Based on the position change trend and real-time movement speed, control commands for the drive components are generated.

7. A monitoring device, applied to the human movement monitoring method as described in any one of claims 1-6, characterized in that, The system includes a processing terminal, a camera, and a driving mechanism for adjusting the angle of the camera. The processing terminal includes: The monitoring image acquisition module (11) is used to acquire real-time monitoring images; Image recognition module (12) is used to perform image recognition processing on the real-time monitoring images; The target following object confirmation module (13) is used to confirm the target following object after the recognition result meets the preset mode switching conditions; The working mode switching module (14) is used to enable the processing terminal to enter the lens following mode after the recognition result meets the preset mode switching conditions; Position change trend recognition module (15) is used to recognize the position change trend of the target following object; The device control module (16) is used to generate control commands for the drive components based on the position change trend, so as to adjust the rotation angle of the camera.

8. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed according to any one of claims 1 to 6.

Images