阅读说明：本技术 课室人员姿态判断方法及装置 (Classroom personnel posture judgment method and device ) 是由 关本立 欧俊文 于 2021-07-20 设计创作，主要内容包括：本发明涉及一种课室人员姿态判断方法及装置,在获取到课室摄像机拍摄的各帧图像,并确定图像的课室人员位置信息后,根据各帧图像中的第一连续帧的课室人员位置信息和像素速度,确定图像的课室人员初始运动信息,并将第二连续帧中每一帧的课室人员位置信息与课室人员初始运动信息进行匹配,得到最新检测结果。进一步地,对最新检测结果进行最新帧预测更新,得到最新帧的预测速度和预测位置,并根据预测速度和预测位置确定课室人员的姿态。基于此,提供一种贴近课室使用场景的方法,准确判断课室人员的姿态变化,为如视频直播或视频录播等在内的控制提供判断基础。(The invention relates to a classroom personnel posture judgment method and device, after acquiring images of frames shot by a classroom camera and determining classroom personnel position information of the images, determining classroom personnel initial motion information of the images according to classroom personnel position information and pixel speed of a first continuous frame in the images of the frames, and matching classroom personnel position information of each frame in a second continuous frame with the classroom personnel initial motion information to obtain a latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.)

1. A classroom personnel posture judgment method is characterized by comprising the following steps:

acquiring each frame of image shot by a classroom camera, and determining classroom personnel position information of the image;

determining the initial motion information of the classroom personnel of the image according to the position information and the pixel speed of the classroom personnel of the first continuous frame in each frame of image;

matching the position information of the classroom staff of each frame in the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result; wherein the second continuous frame is a succeeding continuous frame of the first continuous frame;

carrying out latest frame prediction updating on the latest detection result to obtain the prediction speed and the prediction position of the latest frame; wherein the latest frame is a succeeding frame of the second consecutive frame;

and determining the posture of the classroom staff according to the predicted speed and the predicted position.

2. The classroom staff posture determination method of claim 1, wherein the classroom staff position information includes a position information enclosure.

3. The classroom staff posture determination method of claim 1 wherein said process of determining classroom staff position information for said image comprises the steps of:

and determining classroom staff position information of the image according to a lightweight head detection algorithm.

4. The method for determining the posture of the classroom personnel as described in claim 1, wherein said process of determining the initial motion information of the classroom personnel of said image according to the position information and pixel velocity of the classroom personnel of the first continuous frame in each frame of image comprises the steps of:

and initializing the position information and the pixel speed of the classroom personnel of the first continuous frame by a preset filtering method to obtain the initial motion information of the classroom personnel of the image.

5. The method of claim 1, wherein the process of matching the position information of the classroom staff in each of the second consecutive frames with the initial motion information of the classroom staff to obtain the latest detection result comprises the steps of:

comparing the intersection ratio of the position information of the classroom staff of each frame in the second continuous frames with the intersection ratio of the initial motion information of the classroom staff, and selecting the position information of the classroom staff with the largest intersection ratio and the preset number as candidate position information of the classroom staff;

and taking the position information of the candidate classroom staff closest to the central point of the initial motion information of the classroom staff as the latest detection result.

6. The classroom staff attitude determination method according to claim 1, wherein the process of performing the latest frame prediction update on the latest detection result to obtain the prediction speed and the prediction position of the latest frame comprises the steps of:

and performing latest frame prediction updating on the latest detection result by a preset filtering method to obtain the prediction speed and the prediction position of the latest frame.

7. The classroom personnel posture determination method of claim 4 or 6, wherein the predetermined filtering method comprises Kalman filtering.

8. The classroom staff posture judgment method according to claim 4 or 6, further comprising the steps of:

and when the continuous lost frame number is greater than the preset frame number, controlling the preset filtering method to carry out secondary processing to replace the processing result of the original preset filtering method.

9. The classroom staff attitude determination method of claim 6, wherein the predicted speed includes a horizontal direction speed and a vertical direction speed.

10. The classroom person posture determination method as described in claim 1, wherein said process of determining the posture of the classroom person based on said predicted speed and said predicted position comprises the steps of:

and judging the posture of the classroom staff in the vertical direction according to the size of the preset speed in the vertical direction and the change amplitude of the preset position.

11. The classroom person posture determination method as described in claim 1, wherein said process of determining the posture of the classroom person based on said predicted speed and said predicted position comprises the steps of:

and judging the posture of the classroom staff according to the size of the preset speed in the vertical direction, the size of the preset speed in the horizontal direction and the change frequency of the preset position.

12. The classroom staff pose determination method of claim 1, further comprising the steps of:

and reserving a plurality of frames of the latest frame to update the initial motion information of the classroom personnel.

13. The method for determining the posture of a classroom person as recited in claim 12, wherein said process of maintaining a plurality of said latest frames to update the initial motion information of said classroom person comprises the steps of:

and using the latest frame with the distance between the position information of the classroom staff and the central point of the initial motion information of the classroom staff meeting the matching requirement to update the initial motion information of the classroom staff.

14. A classroom personnel posture judging device, characterized by comprising:

the information acquisition module is used for acquiring each frame of image shot by the classroom camera and determining classroom personnel position information of the image;

the initial determination module is used for determining the initial motion information of the classroom personnel of the image according to the position information and the pixel speed of the classroom personnel of the first continuous frame in each frame of image;

the result matching module is used for matching the position information of the classroom staff of each frame in the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result; wherein the second continuous frame is a succeeding continuous frame of the first continuous frame;

the result prediction module is used for carrying out latest frame prediction updating on the latest detection result to obtain the prediction speed and the prediction position of the latest frame; wherein the latest frame is a succeeding frame of the second consecutive frame;

and the gesture determining module is used for determining the gesture of the classroom staff according to the predicted speed and the predicted position.

15. A computer storage medium having computer instructions stored thereon, wherein the computer instructions, when executed by a processor, implement a classroom personnel pose determination method as defined in any one of claims 1-13.

16. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements a classroom personnel posture assessment method as defined in any one of claims 1 to 13.

Technical Field

The invention relates to the technical field of video teaching, in particular to a method and a device for judging postures of classroom personnel.

Background

With the development of information network technology, education realizes real-time or time-sharing remote education by means of an information network and related matched technical equipment. The remote education mainly depends on an education platform to acquire and summarize teaching data, and each teaching terminal acquires the teaching data through the teaching platform to complete teaching, including video recording and broadcasting teaching or video live broadcasting teaching and the like. The remote education relies on the development of cloud technology, and the set limitations of an education platform and an education terminal, including equipment limitations and user limitations, are further reduced.

In the process of rebroadcasting of video recording and broadcasting teaching or video live broadcasting teaching, a rebroadcasting lens or angle and the like need to be changed according to postures of personnel in a classroom, for example, according to posture changes of students such as standing up and sitting down, a rebroadcasting mode is changed or corresponding control and adjustment are carried out, and the like, so that a live broadcasting strategy or a recording and broadcasting strategy is enriched. There are two common ways to determine the posture change, the first is implemented based on common image processing techniques, such as frame difference method. The main problems of using this technique are susceptible to the interference of the recording environment, such as the brightness of the picture, the change of the background, etc. The other method is based on the common machine learning/deep learning method, and the main method is to train a model capable of detecting a standing human body and then judge the posture change of the classroom personnel by using the model. However, the two schemes do not consider the actual classroom scene, the classroom generally has the sheltering of the desk, the book and other objects, and the whole standing human body cannot be seen in the classroom scene, so that the gesture judgment is lost, and the two methods are mainly applied to the meeting room scene.

In summary, it can be seen that the conventional posture determination method has the above disadvantages.

Disclosure of Invention

Therefore, it is necessary to provide a classroom staff posture determination method and device for the shortcomings of the conventional posture determination method.

A classroom personnel posture judgment method comprises the following steps:

acquiring each frame of image shot by a classroom camera, and determining classroom personnel position information of the image;

determining the initial motion information of the classroom personnel of the image according to the position information and the pixel speed of the classroom personnel of the first continuous frame in each frame of image;

matching the position information of the classroom staff of each frame in the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result; wherein the second continuous frame is a succeeding continuous frame of the first continuous frame;

carrying out latest frame prediction updating on the latest detection result to obtain the prediction speed and the prediction position of the latest frame; wherein the latest frame is a succeeding frame of the second consecutive frame;

and determining the posture of the classroom personnel according to the predicted speed and the predicted position.

According to the method for judging the posture of the classroom staff, after the images of all frames shot by the classroom camera are obtained and the position information of the classroom staff of the images is determined, the initial movement information of the classroom staff of the images is determined according to the position information and the pixel speed of the classroom staff of the first continuous frame in the images of all frames, and the position information of the classroom staff of each frame in the second continuous frame is matched with the initial movement information of the classroom staff to obtain the latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

In one embodiment, the classroom personnel location information includes a location information enclosure.

In one embodiment, the process of determining classroom personnel location information for an image includes the steps of:

and determining classroom staff position information of the image according to a lightweight head detection algorithm.

In one embodiment, the process of determining the initial motion information of the classroom staff of an image according to the position information of the classroom staff and the pixel speed of the first continuous frame in each frame of the image comprises the following steps:

and initializing the position information and the pixel speed of the classroom staff of the first continuous frame by a preset filtering method to obtain the initial motion information of the classroom staff of the image.

In one embodiment, the process of matching the position information of the classroom staff in each frame of the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result comprises the following steps:

comparing the intersection ratio of the position information of the classroom staff of each frame in the second continuous frames with the intersection ratio of the initial motion information of the classroom staff, and selecting the position information of the classroom staff with the largest intersection ratio by a preset number as candidate position information of the classroom staff;

and taking the position information of the candidate classroom staff closest to the central point of the initial motion information of the classroom staff as the latest detection result.

In one embodiment, the process of performing the latest frame prediction update on the latest detection result to obtain the prediction speed and the prediction position of the latest frame includes the steps of:

and performing latest frame prediction updating on the latest detection result by a preset filtering method to obtain the prediction speed and the prediction position of the latest frame.

In one embodiment, the predetermined filtering method includes kalman filtering.

In one embodiment, the method further comprises the following steps:

and when the continuous lost frame number is greater than the preset frame number, controlling the preset filtering method to carry out secondary processing to replace the processing result of the original preset filtering method.

In one embodiment, the predicted speed includes a horizontal direction speed and a vertical direction speed.

In one embodiment, the process of determining the posture of a classroom person based on a predicted speed and predicted position includes the steps of:

and judging the posture of the classroom staff in the vertical direction according to the size of the preset speed in the vertical direction and the change amplitude of the preset position.

In one embodiment, the process of determining the posture of a classroom person based on a predicted speed and predicted position includes the steps of:

and judging the postures of the classroom personnel according to the size of the preset speed in the vertical direction, the size of the preset speed in the horizontal direction and the change frequency of the preset position.

In one embodiment, the method further comprises the following steps:

and reserving a plurality of latest frames to update the initial motion information of the classroom personnel.

In one embodiment, the process of keeping the latest frame of a plurality of frames to update the initial motion information of the classroom personnel comprises the following steps:

and using the latest frame with the distance between the position information of the classroom staff and the central point of the initial motion information of the classroom staff meeting the matching requirement to update the initial motion information of the classroom staff.

A classroom personnel posture judgment device comprising:

the information acquisition module is used for acquiring each frame of image shot by the classroom camera and determining classroom personnel position information of the image;

the initial determination module is used for determining the initial motion information of the classroom personnel of the image according to the position information and the pixel speed of the classroom personnel of the first continuous frame in each frame of image;

the result matching module is used for matching the position information of the classroom staff of each frame in the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result; wherein the second continuous frame is a succeeding continuous frame of the first continuous frame;

the result prediction module is used for carrying out latest frame prediction updating on the latest detection result to obtain the prediction speed and the prediction position of the latest frame; wherein the latest frame is a succeeding frame of the second consecutive frame;

and the gesture determining module is used for determining the gesture of the classroom personnel according to the predicted speed and the predicted position.

After acquiring each frame of image shot by the classroom camera and determining the classroom staff position information of the image, the classroom staff posture judgment device determines the classroom staff initial motion information of the image according to the classroom staff position information and the pixel speed of the first continuous frame in each frame of image, and matches the classroom staff position information of each frame in the second continuous frame with the classroom staff initial motion information to obtain the latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

A computer storage medium having computer instructions stored thereon, the computer instructions when executed by a processor implementing the classroom staff posture assessment method of any of the above embodiments.

After acquiring each frame of image shot by the classroom camera and determining the classroom personnel position information of the image, the computer storage medium determines the classroom personnel initial motion information of the image according to the classroom personnel position information and the pixel speed of the first continuous frame in each frame of image, and matches the classroom personnel position information of each frame in the second continuous frame with the classroom personnel initial motion information to obtain the latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

A computer device comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the classroom personnel posture judgment method of any one of the embodiments.

After the computer equipment acquires each frame of image shot by the classroom camera and determines the classroom personnel position information of the image, the classroom personnel initial motion information of the image is determined according to the classroom personnel position information and the pixel speed of the first continuous frame in each frame of image, and the classroom personnel position information of each frame in the second continuous frame is matched with the classroom personnel initial motion information to obtain the latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

Drawings

FIG. 1 is a flowchart of a classroom personnel posture determination method according to an embodiment;

FIG. 2 is a flowchart of a classroom personnel posture assessment method in accordance with another embodiment;

FIG. 3 is a flowchart of a classroom personnel posture assessment method in accordance with yet another embodiment;

FIG. 4 is a flowchart of a classroom personnel posture assessment method in accordance with yet another embodiment;

FIG. 5 is a block diagram of a classroom personnel posture assessment device according to one embodiment;

FIG. 6 is a schematic diagram of an internal structure of a computer according to an embodiment.

Detailed Description

For better understanding of the objects, technical solutions and effects of the present invention, the present invention will be further explained with reference to the accompanying drawings and examples. Meanwhile, the following described examples are only for explaining the present invention, and are not intended to limit the present invention.

The embodiment of the invention provides a method for judging the posture of a classroom worker.

Fig. 1 is a flowchart of a classroom personnel posture determination method according to an embodiment, and as shown in fig. 1, the classroom personnel posture determination method according to an embodiment includes steps S100 to S104:

s100, acquiring each frame of image shot by a classroom camera, and determining classroom staff position information of the image;

the classroom of the video teaching is provided with a classroom camera, and the classroom camera is used for shooting pictures in a classroom to form images. And capturing corresponding information according to the shot frame images. In one embodiment, the identity of the classroom person is determined by capturing an image of the head of the classroom person. Based on this, the classroom person location information includes a location information enclosure of the classroom person. As a preferred embodiment, the classroom person location information includes a location information bounding box for the head of the classroom person.

In one example, fig. 2 is a flowchart of a classroom person posture determination method according to another embodiment, and as shown in fig. 2, the process of determining classroom person position information of an image in step S100 includes step S200:

and S200, determining classroom staff position information of the image according to a lightweight human head detection algorithm.

The classroom staff position information can be obtained through training, can run in an embedded environment, and is suitable for being applied to a lightweight head detection algorithm in a classroom environment to determine classroom staff position information. Taking images as an example, in the application scene of a classroom, the only dynamic information that has obvious characteristics and is not easily occluded is the head of a classroom person. Meanwhile, for the task of detecting the posture change of the target standing up/sitting down, information of the target in near real time is required, so that the amount of calculation needs to be limited. Generally, the light-weighted detection algorithm is not suitable for detecting more complex targets, such as: for head detection, a light-weight detection algorithm is more suitable in terms of detection accuracy and computation amount.

S101, determining the initial motion information of the classroom personnel of the image according to the position information and the pixel speed of the classroom personnel of the first continuous frame in each frame of image;

and selecting continuous n frames in the image as a first continuous frame, carrying out image detection, and determining the position information of the classroom staff and the corresponding pixel speed. In one embodiment, the classroom person position information is determined in the form of a position information bounding box, and the pixel velocity of the first successive frame includes the pixel velocity of the position information bounding box.

In one embodiment, as shown in fig. 2, the process of determining the initial motion information of the classroom people of the image according to the classroom people position information and the pixel speed of the first continuous frame in each frame of image in step S101 includes the step S201:

s201, initializing the position information and the pixel speed of the classroom staff of the first continuous frame by a preset filtering method, and obtaining the initial motion information of the classroom staff of the image.

And initializing motion data and algorithm internal parameters by a preset filtering method, and determining the initial motion information of the personnel in the class in the first continuous frame. In one embodiment, the predetermined filtering method includes kalman filtering. Due to the image of a single frame, it is difficult to ensure that each classroom person is detected. Thus, missed detection of the classroom personnel is prevented by the first consecutive frame. Meanwhile, since classroom scenes under the first continuous frame are stacked, by initially stacking the motion information of the classroom personnel of the first continuous frame, an approximate motion range of all classroom personnel in the picture under the current environment can be constructed.

In one embodiment, the classroom personnel initial movement information includes a location information bounding box.

S102, matching the position information of the classroom staff of each frame in the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result; wherein the second continuous frame is a succeeding continuous frame of the first continuous frame;

taking the first continuous frame as the 1 st frame to the nth frame as an example, the second continuous frame starts from the (n + 1) th frame, the position information of the classroom personnel of each frame in the two continuous frames is matched with the initial motion information of the classroom personnel, and the motion change of the position information of the classroom personnel is judged.

In one embodiment, as shown in fig. 2, the process of matching the position information of the classroom staff in each frame of the second continuous frames with the initial movement information of the classroom staff in step S102 to obtain the latest detection result includes steps S202 and S203:

s202, comparing the Intersection ratio of the position information of the classroom staff of each frame in the second continuous frames with the Intersection ratio (IOU) of the initial motion information of the classroom staff, and selecting the position information of the classroom staff with the largest Intersection ratio in preset number as candidate classroom staff position information;

and comparing the intersection ratio of the position information of the classroom staff of each frame in the second continuous frames with the intersection ratio of the initial motion information of the classroom staff, and selecting the position information enclosure frame corresponding to the position information of the classroom staff with the largest k IOUs as the position information of the candidate classroom staff.

S203, the position information of the candidate classroom staff closest to the center point of the initial motion information of the classroom staff is used as the latest detection result.

And selecting the candidate classroom staff position information closest to the classroom staff initial motion information central point from the k candidate classroom staff position information as a latest detection result in a central point nearest neighbor mode.

S103, carrying out latest frame prediction updating on the latest detection result to obtain the prediction speed and the prediction position of the latest frame; wherein the latest frame is a succeeding frame of the second consecutive frame;

and performing prediction update on the latest frame according to the latest detection result, and determining the prediction speed and the prediction position of the latest frame. For example, the first continuous frame is the 1 st to nth frame, the second continuous frame is the (n + 1) th to n + m frame, and the latest frame is the (n + m + 1) th frame.

In one embodiment, the predicted speed includes a horizontal direction speed and a vertical direction speed.

Similarly, the predicted position includes a position in the horizontal direction and a position in the vertical direction. In an image, the horizontal direction may be characterized as the x-direction and the vertical direction may be characterized as the y-direction.

In one embodiment, as shown in fig. 2, the step S103 of performing prediction update on the latest frame on the latest detection result to obtain the prediction speed and the prediction position of the latest frame includes the step S204:

and S204, performing latest frame prediction updating on the latest detection result by a preset filtering method to obtain the prediction speed and the prediction position of the latest frame.

The preset filtering method comprises an input and output data observation system including Kalman filtering, and the system state is optimally estimated.

And S104, determining the postures of the classroom personnel according to the predicted speed and the predicted position.

After the prediction speed and the prediction position of the latest frame are determined, the action change of the classroom staff is determined according to the prediction speed and the prediction position so as to determine the posture. In one embodiment, as shown in fig. 2, the process of determining the posture of the classroom person based on the predicted speed and the predicted position in step S104 includes step S205:

and S205, judging the posture of the classroom staff in the vertical direction according to the size of the preset speed in the vertical direction and the change amplitude of the preset position.

When the size of the preset speed in the vertical direction is larger than the first preset speed, the posture of the classroom personnel is considered to be standing up or sitting down. In one embodiment, when the magnitude of the preset speed in the vertical direction is vertically upward, the posture is determined to be standing up. When the magnitude of the preset speed in the vertical direction is vertical downward, the posture is judged to be sitting down.

In one embodiment, fig. 3 is a flowchart of a classroom person posture determination method according to yet another embodiment, and as shown in fig. 3, the process of determining the posture of the classroom person according to the predicted speed and the predicted position in step S104 includes step S300:

and S300, judging the posture of the classroom staff according to the size of the preset speed in the vertical direction, the size of the preset speed in the horizontal direction and the change frequency of the preset position.

When the preset speed is higher than the second preset speed in the vertical direction and the horizontal direction, if the change frequency of the preset position is higher than the preset frequency, the posture of the classroom staff cannot be determined, and the posture is random motion.

Wherein, standing up, sitting down and random movement can be used for guiding the corresponding teaching live broadcasting strategy or recording and broadcasting strategy.

In one embodiment, as shown in fig. 3, the classroom personnel posture judgment method of another embodiment further comprises a step S301:

s301, when the continuous lost frame number is larger than the preset frame number, controlling the preset filtering method to carry out secondary processing to replace the processing result of the original preset filtering method.

And when the frame number larger than the preset frame number is continuously lost, abandoning the original processing result of the preset filtering method, and performing secondary processing to replace the processing result of the original preset filtering method. Based on this, solve the problem that classroom personnel positional information loses or can't obtain.

In one embodiment, as shown in fig. 3, the classroom personnel posture judgment method of another embodiment further comprises a step S302:

s302, reserving a plurality of frames of latest frames to update the initial motion information of the classroom personnel.

In step S302, when the posture of the classroom staff is random, the initial motion information of the classroom staff is updated, so that the initial motion information of the classroom staff can accurately reflect the actual position of the target.

In one embodiment, fig. 4 is a flowchart of a classroom personnel posture determination method according to yet another embodiment, as shown in fig. 4, a process of reserving a plurality of latest frames to update the classroom personnel initial movement information in step S302 includes step S400:

s400, the latest frame with the distance between the position information of the classroom staff and the central point of the initial motion information of the classroom staff meeting the matching requirement is used for updating the initial motion information of the classroom staff.

The distance between the classroom staff position information and the classroom staff initial movement information center point meets the matching requirement, namely the distance between the classroom staff position information and the classroom staff initial movement information center point is smaller than a preset distance threshold value.

In the above classroom staff posture judgment method of any embodiment, after acquiring each frame of image shot by the classroom camera and determining the classroom staff position information of the image, according to the classroom staff position information and the pixel speed of the first continuous frame in each frame of image, the classroom staff initial motion information of the image is determined, and the classroom staff position information of each frame in the second continuous frame is matched with the classroom staff initial motion information to obtain the latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

The embodiment of the invention also provides a device for judging the posture of the classroom personnel.

Fig. 5 is a block diagram of a classroom personnel posture determination device according to an embodiment, and as shown in fig. 5, the classroom personnel posture determination device according to an embodiment includes a module 100, a module 101, a module 102, a module 103, and a module 104:

an information acquisition module 100, configured to acquire each frame of image captured by a classroom camera, and determine classroom staff position information of the image;

an initial determination module 101, configured to determine, according to the position information and the pixel speed of the classroom staff of the first continuous frame in each frame of image, classroom staff initial movement information of the image;

the result matching module 102 is used for matching the position information of the classroom staff of each frame in the second continuous frames with the initial motion information of the classroom staff to obtain the latest detection result; wherein the second continuous frame is a succeeding continuous frame of the first continuous frame;

the result prediction module 103 is configured to perform latest frame prediction update on the latest detection result to obtain a prediction speed and a prediction position of the latest frame; wherein the latest frame is a succeeding frame of the second consecutive frame;

and the gesture determining module 104 is used for determining the gesture of the classroom personnel according to the predicted speed and the predicted position.

After acquiring each frame of image shot by the classroom camera and determining the classroom staff position information of the image, the classroom staff posture judgment device determines the classroom staff initial motion information of the image according to the classroom staff position information and the pixel speed of the first continuous frame in each frame of image, and matches the classroom staff position information of each frame in the second continuous frame with the classroom staff initial motion information to obtain the latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

The embodiment of the invention also provides a computer storage medium, wherein computer instructions are stored on the computer storage medium, and when the instructions are executed by a processor, the method for judging the posture of the classroom personnel in any embodiment is realized.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a terminal, or a network device) to execute all or part of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a RAM, a ROM, a magnetic or optical disk, or various other media that can store program code.

Corresponding to the computer storage medium, in one embodiment, a computer device is further provided, the computer device includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement any one of the above-mentioned methods for determining the posture of the classroom person.

The computer device may be a terminal, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a classroom staff posture determination method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

After acquiring each frame of image shot by the classroom camera and determining classroom staff position information of the image, the computer equipment determines classroom staff initial motion information of the image according to classroom staff position information and pixel speed of a first continuous frame in each frame of image, and matches classroom staff position information of each frame in a second continuous frame with classroom staff initial motion information to obtain a latest detection result. Furthermore, the latest frame prediction updating is carried out on the latest detection result to obtain the prediction speed and the prediction position of the latest frame, and the posture of the classroom staff is determined according to the prediction speed and the prediction position. Therefore, the method close to the classroom use scene is provided, the posture change of classroom personnel is accurately judged, and a judgment basis is provided for control such as live video broadcast or recorded video broadcast.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above examples only show some embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.